* [PATCH V3] arm64: dts: ls1046a: Add TMU device tree support
From: Jia Hongtao @ 2017-01-04 2:36 UTC (permalink / raw)
To: linux-arm-kernel
Also add nodes and properties for thermal management support.
Signed-off-by: Jia Hongtao <hongtao.jia@nxp.com>
---
Changes for V3:
* Update the subject title according to Shawn Guo's comment.
* Fix some style issue.
Changes for V2:
* Update the subject title according to Shawn Guo's comment.
* Add comments for calibration data groups.
* Update "thermal-zones" property in a unified style with platform dts.
arch/arm64/boot/dts/freescale/fsl-ls1046a.dtsi | 80 ++++++++++++++++++++++++++
1 file changed, 80 insertions(+)
diff --git a/arch/arm64/boot/dts/freescale/fsl-ls1046a.dtsi b/arch/arm64/boot/dts/freescale/fsl-ls1046a.dtsi
index 38806ca..4a164b8 100644
--- a/arch/arm64/boot/dts/freescale/fsl-ls1046a.dtsi
+++ b/arch/arm64/boot/dts/freescale/fsl-ls1046a.dtsi
@@ -45,6 +45,7 @@
*/
#include <dt-bindings/interrupt-controller/arm-gic.h>
+#include <dt-bindings/thermal/thermal.h>
/ {
compatible = "fsl,ls1046a";
@@ -67,6 +68,7 @@
clocks = <&clockgen 1 0>;
next-level-cache = <&l2>;
cpu-idle-states = <&CPU_PH20>;
+ #cooling-cells = <2>;
};
cpu1: cpu at 1 {
@@ -279,6 +281,84 @@
clocks = <&sysclk>;
};
+ tmu: tmu at 1f00000 {
+ compatible = "fsl,qoriq-tmu";
+ reg = <0x0 0x1f00000 0x0 0x10000>;
+ interrupts = <0 33 0x4>;
+ fsl,tmu-range = <0xb0000 0x9002a 0x6004c 0x30062>;
+ fsl,tmu-calibration =
+ /* Calibration data group 1 */
+ <0x00000000 0x00000026
+ 0x00000001 0x0000002d
+ 0x00000002 0x00000032
+ 0x00000003 0x00000039
+ 0x00000004 0x0000003f
+ 0x00000005 0x00000046
+ 0x00000006 0x0000004d
+ 0x00000007 0x00000054
+ 0x00000008 0x0000005a
+ 0x00000009 0x00000061
+ 0x0000000a 0x0000006a
+ 0x0000000b 0x00000071
+ /* Calibration data group 2 */
+ 0x00010000 0x00000025
+ 0x00010001 0x0000002c
+ 0x00010002 0x00000035
+ 0x00010003 0x0000003d
+ 0x00010004 0x00000045
+ 0x00010005 0x0000004e
+ 0x00010006 0x00000057
+ 0x00010007 0x00000061
+ 0x00010008 0x0000006b
+ 0x00010009 0x00000076
+ /* Calibration data group 3 */
+ 0x00020000 0x00000029
+ 0x00020001 0x00000033
+ 0x00020002 0x0000003d
+ 0x00020003 0x00000049
+ 0x00020004 0x00000056
+ 0x00020005 0x00000061
+ 0x00020006 0x0000006d
+ /* Calibration data group 4 */
+ 0x00030000 0x00000021
+ 0x00030001 0x0000002a
+ 0x00030002 0x0000003c
+ 0x00030003 0x0000004e>;
+ big-endian;
+ #thermal-sensor-cells = <1>;
+ };
+
+ thermal-zones {
+ cpu_thermal: cpu-thermal {
+ polling-delay-passive = <1000>;
+ polling-delay = <5000>;
+ thermal-sensors = <&tmu 3>;
+
+ trips {
+ cpu_alert: cpu-alert {
+ temperature = <85000>;
+ hysteresis = <2000>;
+ type = "passive";
+ };
+
+ cpu_crit: cpu-crit {
+ temperature = <95000>;
+ hysteresis = <2000>;
+ type = "critical";
+ };
+ };
+
+ cooling-maps {
+ map0 {
+ trip = <&cpu_alert>;
+ cooling-device =
+ <&cpu0 THERMAL_NO_LIMIT
+ THERMAL_NO_LIMIT>;
+ };
+ };
+ };
+ };
+
dspi: dspi at 2100000 {
compatible = "fsl,ls1021a-v1.0-dspi";
#address-cells = <1>;
--
2.1.0.27.g96db324
^ permalink raw reply related
* [PATCH 3/5] arm64: dts: sun50i: add MMC nodes
From: Chen-Yu Tsai @ 2017-01-04 2:37 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <6e90ad76-8f25-538a-433e-ab31e8fab526@arm.com>
On Wed, Jan 4, 2017 at 8:22 AM, Andr? Przywara <andre.przywara@arm.com> wrote:
> On 03/01/17 13:28, Chen-Yu Tsai wrote:
>> On Tue, Jan 3, 2017 at 6:48 PM, Andr? Przywara <andre.przywara@arm.com> wrote:
>>> On 03/01/17 02:52, Chen-Yu Tsai wrote:
>
> Hi Chen-Yu,
>
>>>> On Tue, Jan 3, 2017 at 7:03 AM, Andre Przywara <andre.przywara@arm.com> wrote:
>>>>
>>>> A commit message explaining the mmc controllers would be nice.
>>>
>>> OK.
>>>
>>>>> Signed-off-by: Andre Przywara <andre.przywara@arm.com>
>>>>> ---
>>>>> arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi | 67 +++++++++++++++++++++++++++
>>>>> 1 file changed, 67 insertions(+)
>>>>>
>>>>> diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
>>>>> index e0dcab8..c680566 100644
>>>>> --- a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
>>>>> +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
>>>>> @@ -150,6 +150,32 @@
>>>>> pins = "PB8", "PB9";
>>>>> function = "uart0";
>>>>> };
>>>>> +
>>>>> + mmc0_pins: mmc0 at 0 {
>>>>> + pins = "PF0", "PF1", "PF2", "PF3", "PF4", "PF5";
>>>>> + function = "mmc0";
>>>>> + drive-strength = <30>;
>>>>> + };
>>>>> +
>>>>> + mmc0_default_cd_pin: mmc0_cd_pin at 0 {
>>>>> + pins = "PF6";
>>>>> + function = "gpio_in";
>>>>> + bias-pull-up;
>>>>> + };
>>>>
>>>> We are starting to drop pinmux nodes for gpio usage.
>>>
>>> And replacing them with what?
>>> Or do you mean they go in the individual board .dts files?
>>> In this case I believe having a default pin defined here would help to
>>> define it in every .dts.
>>
>> Nope. I meant dropping them. Pinmux and gpio are orthogonal. One should not
>> need to specify a gpio pinmux to use it as a gpio. We added them because in
>> the past nothing was preventing someone from claiming an already muxed pin
>> as a gpio. On some platforms this is fine. For sunxi, this breaks the system,
>> as the gpio functions are muxed in.
>>
>> The idea moving forward is that these cases should be guarded in the driver.
>
> Ah, OK, you mean just referencing the pin like:
> cd-gpios = <&pio 5 6 GPIO_ACTIVE_HIGH>; /* PF6 */
> in the board DT is enough? And the driver claiming it will (eventually?)
> prevent users from using them via the sysfs interface then?
cd-gpios = ....
should be enough to block other devices from claiming a pinmux (pinctrl-N = ...)
on the same pin, or vice versa, whichever claimed it first in kernel space.
Same goes for a pin already claimed by a device through a pinmux. It should
not be claimable as a gpio through the sysfs interface.
>> Of course we would have to deal with existing dtbs, but lets not add any more.
>>
>>>>> +
>>>>> + mmc1_pins: mmc1 at 0 {
>>>>> + pins = "PG0", "PG1", "PG2", "PG3", "PG4", "PG5";
>>>>> + function = "mmc1";
>>>>> + drive-strength = <30>;
>>>>> + };
>>>>> +
>>>>> + mmc2_pins: mmc2 at 0 {
>>>>> + pins = "PC1", "PC5", "PC6", "PC8", "PC9",
>>>>> + "PC10", "PC11", "PC12", "PC13", "PC14",
>>>>> + "PC15", "PC16";
>>>>> + function = "mmc2";
>>>>> + drive-strength = <30>;
>>>>> + };
>>>>
>>>> Moreover I think you should split out the pinmux nodes to a separate patch.
>>>
>>> I can surely do, just wondering what's the rationale is behind that?
>>
>> More or less the "do one thing in one patch" rationale. Of course you can
>> claim these are the defaults used in the reference design and pretty much
>> every board out there. Then it makes sense to do them together. :)
>
> Mmmh, probably a misunderstanding, but:
> Those pins are the possible pins that expose the MMC interface. Those
> nodes here name them to allow easy and concise reference by a board DT
> (as in the following patch).
> And in fact in case of the A64 there are _no_ alternative muxes for the
> three MMC controllers: SDC0 is only on PF0-5, SDC1 on PG0-PG5 and SDC2
> on PC1-PC16.
> So it's not a board design question, apart from using a controller or not.
>
> That's why I rather keep them together with the MMC controller nodes.
Cool. Please mention this in the commit log, as in they are the only
possible choice.S ince they are the only choices, please drop the @0
in the node name.
And you might want to rename mmc2_pins to something like
"mmc2_8bit_pins: mmc2_8bit". Who knows, someone might consider using
it for an SD card or SDIO, and the remaining pins for something else.
>>>>
>>>>> };
>>>>>
>>>>> uart0: serial at 1c28000 {
>>>>> @@ -240,6 +266,47 @@
>>>>> #size-cells = <0>;
>>>>> };
>>>>>
>>>>> + mmc0: mmc at 1c0f000 {
>>>>> + compatible = "allwinner,sun50i-a64-mmc",
>>>>> + "allwinner,sun5i-a13-mmc";
>>>>
>>>> Given that sun5i doesn't support mmc delay timings, and the A64 has
>>>> calibration and delay timings, I wouldn't call them compatible.
>>>>
>>>> Or are you claiming that for the A64 has a delay of 0 for the
>>>> currently supported speeds, so the calibration doesn't really
>>>> matter? If so this should be mentioned in the commit message.
>>>
>>> Yes, that's my observation: Driving it with sun5-a13-mmc just works.
>>> This sun5i driver version does not (and will never) support higher
>>> transfer modes anyway, so for that subset they are compatible. This
>>> opens up the door to other operating systems not having a particular
>>> driver for the A64, for instance, also older Linux kernels.
>>> I know that sunxi doesn't use this compatible feature much, but IMHO we
>>> should really start thinking about the DT not just being Linux specific
>>> - or even being specific to a certain Linux version. And this case here
>>> is a good example: An A13 MMC driver can drive this device - if there is
>>> no better driver (an A64 one) available.
>>
>> Cool. Please put this in the commit log. :)
>
> Noted for a repost.
>
>>>
>>>>
>>>>> + reg = <0x01c0f000 0x1000>;
>>>>> + clocks = <&ccu 31>, <&ccu 75>;
>>>>
>>>> The clock / reset index macros are in the tree now.
>>>> Please switch to them.
>>>
>>> The include file is in the tree, but it isn't used in the current HEAD
>>> (as in #included and the actual macros being used in the .dtsi).
>>> So I was wondering if there is a patch pending for that?
>>
>> Not yet I think. Perhaps Maxime will do one once he gets back from vacation?
>
> Seems like everybody hopes for it and nobody bothers with this rather
> trivial patch ;-)
A bit of context here. This was changed _after_ the A64 patches were
pulled into arm-soc, by the arm-soc maintainers, to avoid ugly cross
tree merges. The idea was to change them back after -rc1. I don't know
if a patch to do that is ready though.
> Frankly I am not overly keen on having all those defines in the DTs,
> which rely on a bunch of include files scattered over the Linux include
> tree. This makes it unnecessarily complicated to compile this out of
> tree or move it over to some other piece of software (other OS, U-Boot).
> Also it gives people the impression that it's fine to change the
> definition in the include file at will (because it looks like Linux
> territory).
The dt-bindings related header files are split into a separate directory.
They are also included in the separate devicetree repository:
https://git.kernel.org/cgit/linux/kernel/git/devicetree/devicetree-rebasing.git/
And no, I don't think it implies that it's fine to change it. The
include/dt-bindings/ subdirectory is part of the device tree bindings,
just like Documentation/devicetree/bindings. Does the latter give the
impression that it is fine to change it, because it's in the docs?
> Eventually some file has to hold the actual hardware information, I
> don't see why this shouldn't be the one .dtsi file. Reading that the
> clock for MMC controller 0 is MMC0_CLK isn't very informative.
For me, it makes it much easier to review, than say a plain number,
which I would have to dig through some file, probably the clock driver,
to check if it matches.
Regards
ChenYu
> Cheers,
> Andre.
>
>>>>> + clock-names = "ahb", "mmc";
>>>>> + resets = <&ccu 8>;
>>>>> + reset-names = "ahb";
>>>>> + interrupts = <GIC_SPI 60 IRQ_TYPE_LEVEL_HIGH>;
>>>>> + status = "disabled";
>>>>> + #address-cells = <1>;
>>>>> + #size-cells = <0>;
>>>>> + };
>>>>> +
>>>>> + mmc1: mmc at 1c10000 {
>>>>> + compatible = "allwinner,sun50i-a64-mmc",
>>>>> + "allwinner,sun5i-a13-mmc";
>>>>> + reg = <0x01c10000 0x1000>;
>>>>> + clocks = <&ccu 32>, <&ccu 76>;
>>>>> + clock-names = "ahb", "mmc";
>>>>> + resets = <&ccu 9>;
>>>>> + reset-names = "ahb";
>>>>> + interrupts = <GIC_SPI 61 IRQ_TYPE_LEVEL_HIGH>;
>>>>> + status = "disabled";
>>>>> + #address-cells = <1>;
>>>>> + #size-cells = <0>;
>>>>> + };
>>>>> +
>>>>> + mmc2: mmc at 1c11000 {
>>>>> + compatible = "allwinner,sun50i-a64-emmc";
>>>>> + reg = <0x01c11000 0x1000>;
>>>>> + clocks = <&ccu 33>, <&ccu 77>;
>>>>> + clock-names = "ahb", "mmc";
>>>>> + resets = <&ccu 10>;
>>>>> + reset-names = "ahb";
>>>>> + interrupts = <GIC_SPI 62 IRQ_TYPE_LEVEL_HIGH>;
>>>>> + status = "disabled";
>>>>> + #address-cells = <1>;
>>>>> + #size-cells = <0>;
>>>>> + };
>>>>> +
>>>>> gic: interrupt-controller at 1c81000 {
>>>>> compatible = "arm,gic-400";
>>>>> reg = <0x01c81000 0x1000>,
>>>>> --
>>>>> 2.8.2
>>>>>
>>>
>
^ permalink raw reply
* [PATCH v20 0/4] Mediatek MT8173 CMDQ support
From: HS Liao @ 2017-01-04 3:06 UTC (permalink / raw)
To: linux-arm-kernel
Hi,
This is Mediatek MT8173 Command Queue(CMDQ) driver. The CMDQ is used
to help write registers with critical time limitation, such as
updating display configuration during the vblank. It controls Global
Command Engine (GCE) hardware to achieve this requirement.
These patches have a build dependency on top of v4.10-rc2.
Changes since v19:
- rebase to v4.10-rc2
Best regards,
HS Liao
HS Liao (4):
dt-bindings: soc: Add documentation for the MediaTek GCE unit
mailbox: mediatek: Add Mediatek CMDQ driver
arm64: dts: mt8173: Add GCE node
soc: mediatek: Add Mediatek CMDQ helper
.../devicetree/bindings/mailbox/mtk-gce.txt | 43 ++
arch/arm64/boot/dts/mediatek/mt8173.dtsi | 10 +
drivers/mailbox/Kconfig | 10 +
drivers/mailbox/Makefile | 2 +
drivers/mailbox/mtk-cmdq-mailbox.c | 596 +++++++++++++++++++++
drivers/soc/mediatek/Kconfig | 12 +
drivers/soc/mediatek/Makefile | 1 +
drivers/soc/mediatek/mtk-cmdq-helper.c | 310 +++++++++++
include/linux/mailbox/mtk-cmdq-mailbox.h | 75 +++
include/linux/soc/mediatek/mtk-cmdq.h | 174 ++++++
10 files changed, 1233 insertions(+)
create mode 100644 Documentation/devicetree/bindings/mailbox/mtk-gce.txt
create mode 100644 drivers/mailbox/mtk-cmdq-mailbox.c
create mode 100644 drivers/soc/mediatek/mtk-cmdq-helper.c
create mode 100644 include/linux/mailbox/mtk-cmdq-mailbox.h
create mode 100644 include/linux/soc/mediatek/mtk-cmdq.h
--
1.9.1
^ permalink raw reply
* [PATCH v20 1/4] dt-bindings: soc: Add documentation for the MediaTek GCE unit
From: HS Liao @ 2017-01-04 3:06 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483499169-16329-1-git-send-email-hs.liao@mediatek.com>
This adds documentation for the MediaTek Global Command Engine (GCE) unit
found in MT8173 SoCs.
Signed-off-by: HS Liao <hs.liao@mediatek.com>
Acked-by: Rob Herring <robh@kernel.org>
---
.../devicetree/bindings/mailbox/mtk-gce.txt | 43 ++++++++++++++++++++++
1 file changed, 43 insertions(+)
create mode 100644 Documentation/devicetree/bindings/mailbox/mtk-gce.txt
diff --git a/Documentation/devicetree/bindings/mailbox/mtk-gce.txt b/Documentation/devicetree/bindings/mailbox/mtk-gce.txt
new file mode 100644
index 0000000..d2d3ccb
--- /dev/null
+++ b/Documentation/devicetree/bindings/mailbox/mtk-gce.txt
@@ -0,0 +1,43 @@
+MediaTek GCE
+===============
+
+The Global Command Engine (GCE) is used to help read/write registers with
+critical time limitation, such as updating display configuration during the
+vblank. The GCE can be used to implement the Command Queue (CMDQ) driver.
+
+CMDQ driver uses mailbox framework for communication. Please refer to
+mailbox.txt for generic information about mailbox device-tree bindings.
+
+Required properties:
+- compatible: Must be "mediatek,mt8173-gce"
+- reg: Address range of the GCE unit
+- interrupts: The interrupt signal from the GCE block
+- clock: Clocks according to the common clock binding
+- clock-names: Must be "gce" to stand for GCE clock
+- #mbox-cells: Should be 2
+
+Required properties for a client device:
+- mboxes: client use mailbox to communicate with GCE, it should have this
+ property and list of phandle, mailbox channel specifiers, and atomic
+ execution flag.
+
+Example:
+
+ gce: gce at 10212000 {
+ compatible = "mediatek,mt8173-gce";
+ reg = <0 0x10212000 0 0x1000>;
+ interrupts = <GIC_SPI 135 IRQ_TYPE_LEVEL_LOW>;
+ clocks = <&infracfg CLK_INFRA_GCE>;
+ clock-names = "gce";
+
+ #mbox-cells = <2>;
+ };
+
+Example for a client device:
+
+ mmsys: clock-controller at 14000000 {
+ compatible = "mediatek,mt8173-mmsys";
+ mboxes = <&gce 0 1 /* main display with atomic execution */
+ &gce 1 1>; /* sub display with atomic execution */
+ ...
+ };
--
1.9.1
^ permalink raw reply related
* [PATCH v20 2/4] mailbox: mediatek: Add Mediatek CMDQ driver
From: HS Liao @ 2017-01-04 3:06 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483499169-16329-1-git-send-email-hs.liao@mediatek.com>
This patch is first version of Mediatek Command Queue(CMDQ) driver. The
CMDQ is used to help write registers with critical time limitation,
such as updating display configuration during the vblank. It controls
Global Command Engine (GCE) hardware to achieve this requirement.
Currently, CMDQ only supports display related hardwares, but we expect
it can be extended to other hardwares for future requirements.
Signed-off-by: HS Liao <hs.liao@mediatek.com>
Signed-off-by: CK Hu <ck.hu@mediatek.com>
---
drivers/mailbox/Kconfig | 10 +
drivers/mailbox/Makefile | 2 +
drivers/mailbox/mtk-cmdq-mailbox.c | 596 +++++++++++++++++++++++++++++++
include/linux/mailbox/mtk-cmdq-mailbox.h | 75 ++++
4 files changed, 683 insertions(+)
create mode 100644 drivers/mailbox/mtk-cmdq-mailbox.c
create mode 100644 include/linux/mailbox/mtk-cmdq-mailbox.h
diff --git a/drivers/mailbox/Kconfig b/drivers/mailbox/Kconfig
index ceff415..9108dd4 100644
--- a/drivers/mailbox/Kconfig
+++ b/drivers/mailbox/Kconfig
@@ -152,4 +152,14 @@ config BCM_PDC_MBOX
Mailbox implementation for the Broadcom PDC ring manager,
which provides access to various offload engines on Broadcom
SoCs. Say Y here if you want to use the Broadcom PDC.
+
+config MTK_CMDQ_MBOX
+ bool "MediaTek CMDQ Mailbox Support"
+ depends on ARM64 && ( ARCH_MEDIATEK || COMPILE_TEST )
+ select MTK_INFRACFG
+ help
+ Say yes here to add support for the MediaTek Command Queue (CMDQ)
+ mailbox driver. The CMDQ is used to help read/write registers with
+ critical time limitation, such as updating display configuration
+ during the vblank.
endif
diff --git a/drivers/mailbox/Makefile b/drivers/mailbox/Makefile
index 7dde4f6..fad8965 100644
--- a/drivers/mailbox/Makefile
+++ b/drivers/mailbox/Makefile
@@ -31,3 +31,5 @@ obj-$(CONFIG_HI6220_MBOX) += hi6220-mailbox.o
obj-$(CONFIG_BCM_PDC_MBOX) += bcm-pdc-mailbox.o
obj-$(CONFIG_TEGRA_HSP_MBOX) += tegra-hsp.o
+
+obj-$(CONFIG_MTK_CMDQ_MBOX) += mtk-cmdq-mailbox.o
diff --git a/drivers/mailbox/mtk-cmdq-mailbox.c b/drivers/mailbox/mtk-cmdq-mailbox.c
new file mode 100644
index 0000000..747bcd3
--- /dev/null
+++ b/drivers/mailbox/mtk-cmdq-mailbox.c
@@ -0,0 +1,596 @@
+/*
+ * Copyright (c) 2015 MediaTek Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/bitops.h>
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/dma-mapping.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/iopoll.h>
+#include <linux/kernel.h>
+#include <linux/mailbox_controller.h>
+#include <linux/mailbox/mtk-cmdq-mailbox.h>
+#include <linux/timer.h>
+
+#define CMDQ_THR_MAX_COUNT 3 /* main, sub, general(misc) */
+#define CMDQ_OP_CODE_MASK (0xff << CMDQ_OP_CODE_SHIFT)
+#define CMDQ_TIMEOUT_MS 1000
+#define CMDQ_IRQ_MASK 0xffff
+#define CMDQ_NUM_CMD(t) (t->cmd_buf_size / CMDQ_INST_SIZE)
+
+#define CMDQ_CURR_IRQ_STATUS 0x10
+#define CMDQ_THR_SLOT_CYCLES 0x30
+
+#define CMDQ_THR_BASE 0x100
+#define CMDQ_THR_SIZE 0x80
+#define CMDQ_THR_WARM_RESET 0x00
+#define CMDQ_THR_ENABLE_TASK 0x04
+#define CMDQ_THR_SUSPEND_TASK 0x08
+#define CMDQ_THR_CURR_STATUS 0x0c
+#define CMDQ_THR_IRQ_STATUS 0x10
+#define CMDQ_THR_IRQ_ENABLE 0x14
+#define CMDQ_THR_CURR_ADDR 0x20
+#define CMDQ_THR_END_ADDR 0x24
+#define CMDQ_THR_WAIT_TOKEN 0x30
+
+#define CMDQ_THR_ENABLED 0x1
+#define CMDQ_THR_DISABLED 0x0
+#define CMDQ_THR_SUSPEND 0x1
+#define CMDQ_THR_RESUME 0x0
+#define CMDQ_THR_STATUS_SUSPENDED BIT(1)
+#define CMDQ_THR_DO_WARM_RESET BIT(0)
+#define CMDQ_THR_ACTIVE_SLOT_CYCLES 0x3200
+#define CMDQ_THR_IRQ_DONE 0x1
+#define CMDQ_THR_IRQ_ERROR 0x12
+#define CMDQ_THR_IRQ_EN (CMDQ_THR_IRQ_ERROR | CMDQ_THR_IRQ_DONE)
+#define CMDQ_THR_IS_WAITING BIT(31)
+
+#define CMDQ_JUMP_BY_OFFSET 0x10000000
+#define CMDQ_JUMP_BY_PA 0x10000001
+
+struct cmdq_thread {
+ struct mbox_chan *chan;
+ void __iomem *base;
+ struct list_head task_busy_list;
+ struct timer_list timeout;
+ bool atomic_exec;
+};
+
+struct cmdq_task {
+ struct cmdq *cmdq;
+ struct list_head list_entry;
+ dma_addr_t pa_base;
+ struct cmdq_thread *thread;
+ struct cmdq_pkt *pkt; /* the packet sent from mailbox client */
+};
+
+struct cmdq {
+ struct mbox_controller mbox;
+ void __iomem *base;
+ u32 irq;
+ struct cmdq_thread thread[CMDQ_THR_MAX_COUNT];
+ struct clk *clock;
+ bool suspended;
+};
+
+static int cmdq_thread_suspend(struct cmdq *cmdq, struct cmdq_thread *thread)
+{
+ u32 status;
+
+ writel(CMDQ_THR_SUSPEND, thread->base + CMDQ_THR_SUSPEND_TASK);
+
+ /* If already disabled, treat as suspended successful. */
+ if (!(readl(thread->base + CMDQ_THR_ENABLE_TASK) & CMDQ_THR_ENABLED))
+ return 0;
+
+ if (readl_poll_timeout_atomic(thread->base + CMDQ_THR_CURR_STATUS,
+ status, status & CMDQ_THR_STATUS_SUSPENDED, 0, 10)) {
+ dev_err(cmdq->mbox.dev, "suspend GCE thread 0x%x failed\n",
+ (u32)(thread->base - cmdq->base));
+ return -EFAULT;
+ }
+
+ return 0;
+}
+
+static void cmdq_thread_resume(struct cmdq_thread *thread)
+{
+ writel(CMDQ_THR_RESUME, thread->base + CMDQ_THR_SUSPEND_TASK);
+}
+
+static int cmdq_thread_reset(struct cmdq *cmdq, struct cmdq_thread *thread)
+{
+ u32 warm_reset;
+
+ writel(CMDQ_THR_DO_WARM_RESET, thread->base + CMDQ_THR_WARM_RESET);
+ if (readl_poll_timeout_atomic(thread->base + CMDQ_THR_WARM_RESET,
+ warm_reset, !(warm_reset & CMDQ_THR_DO_WARM_RESET),
+ 0, 10)) {
+ dev_err(cmdq->mbox.dev, "reset GCE thread 0x%x failed\n",
+ (u32)(thread->base - cmdq->base));
+ return -EFAULT;
+ }
+ writel(CMDQ_THR_ACTIVE_SLOT_CYCLES, cmdq->base + CMDQ_THR_SLOT_CYCLES);
+ return 0;
+}
+
+static void cmdq_thread_disable(struct cmdq *cmdq, struct cmdq_thread *thread)
+{
+ cmdq_thread_reset(cmdq, thread);
+ writel(CMDQ_THR_DISABLED, thread->base + CMDQ_THR_ENABLE_TASK);
+}
+
+/* notify GCE to re-fetch commands by setting GCE thread PC */
+static void cmdq_thread_invalidate_fetched_data(struct cmdq_thread *thread)
+{
+ writel(readl(thread->base + CMDQ_THR_CURR_ADDR),
+ thread->base + CMDQ_THR_CURR_ADDR);
+}
+
+static void cmdq_task_insert_into_thread(struct cmdq_task *task)
+{
+ struct device *dev = task->cmdq->mbox.dev;
+ struct cmdq_thread *thread = task->thread;
+ struct cmdq_task *prev_task = list_last_entry(
+ &thread->task_busy_list, typeof(*task), list_entry);
+ u64 *prev_task_base = prev_task->pkt->va_base;
+
+ /* let previous task jump to this task */
+ dma_sync_single_for_cpu(dev, prev_task->pa_base,
+ prev_task->pkt->cmd_buf_size, DMA_TO_DEVICE);
+ prev_task_base[CMDQ_NUM_CMD(prev_task->pkt) - 1] =
+ (u64)CMDQ_JUMP_BY_PA << 32 | task->pa_base;
+ dma_sync_single_for_device(dev, prev_task->pa_base,
+ prev_task->pkt->cmd_buf_size, DMA_TO_DEVICE);
+
+ cmdq_thread_invalidate_fetched_data(thread);
+}
+
+static bool cmdq_command_is_wfe(u64 cmd)
+{
+ u64 wfe_option = CMDQ_WFE_UPDATE | CMDQ_WFE_WAIT | CMDQ_WFE_WAIT_VALUE;
+ u64 wfe_op = (u64)(CMDQ_CODE_WFE << CMDQ_OP_CODE_SHIFT) << 32;
+ u64 wfe_mask = (u64)CMDQ_OP_CODE_MASK << 32 | 0xffffffff;
+
+ return ((cmd & wfe_mask) == (wfe_op | wfe_option));
+}
+
+/* we assume tasks in the same display GCE thread are waiting the same event. */
+static void cmdq_task_remove_wfe(struct cmdq_task *task)
+{
+ struct device *dev = task->cmdq->mbox.dev;
+ u64 *base = task->pkt->va_base;
+ int i;
+
+ dma_sync_single_for_cpu(dev, task->pa_base, task->pkt->cmd_buf_size,
+ DMA_TO_DEVICE);
+ for (i = 0; i < CMDQ_NUM_CMD(task->pkt); i++)
+ if (cmdq_command_is_wfe(base[i]))
+ base[i] = (u64)CMDQ_JUMP_BY_OFFSET << 32 |
+ CMDQ_JUMP_PASS;
+ dma_sync_single_for_device(dev, task->pa_base, task->pkt->cmd_buf_size,
+ DMA_TO_DEVICE);
+}
+
+static bool cmdq_thread_is_in_wfe(struct cmdq_thread *thread)
+{
+ return readl(thread->base + CMDQ_THR_WAIT_TOKEN) & CMDQ_THR_IS_WAITING;
+}
+
+static void cmdq_thread_wait_end(struct cmdq_thread *thread,
+ unsigned long end_pa)
+{
+ struct device *dev = thread->chan->mbox->dev;
+ unsigned long curr_pa;
+
+ if (readl_poll_timeout_atomic(thread->base + CMDQ_THR_CURR_ADDR,
+ curr_pa, curr_pa == end_pa, 1, 20))
+ dev_err(dev, "GCE thread cannot run to end.\n");
+}
+
+static void cmdq_task_exec(struct cmdq_pkt *pkt, struct cmdq_thread *thread)
+{
+ struct cmdq *cmdq;
+ struct cmdq_task *task;
+ unsigned long curr_pa, end_pa;
+
+ cmdq = dev_get_drvdata(thread->chan->mbox->dev);
+
+ /* Client should not flush new tasks if suspended. */
+ WARN_ON(cmdq->suspended);
+
+ task = kzalloc(sizeof(*task), GFP_ATOMIC);
+ task->cmdq = cmdq;
+ INIT_LIST_HEAD(&task->list_entry);
+ task->pa_base = dma_map_single(cmdq->mbox.dev, pkt->va_base,
+ pkt->cmd_buf_size, DMA_TO_DEVICE);
+ task->thread = thread;
+ task->pkt = pkt;
+
+ if (list_empty(&thread->task_busy_list)) {
+ WARN_ON(clk_enable(cmdq->clock) < 0);
+ WARN_ON(cmdq_thread_reset(cmdq, thread) < 0);
+
+ writel(task->pa_base, thread->base + CMDQ_THR_CURR_ADDR);
+ writel(task->pa_base + pkt->cmd_buf_size,
+ thread->base + CMDQ_THR_END_ADDR);
+ writel(CMDQ_THR_IRQ_EN, thread->base + CMDQ_THR_IRQ_ENABLE);
+ writel(CMDQ_THR_ENABLED, thread->base + CMDQ_THR_ENABLE_TASK);
+
+ mod_timer(&thread->timeout,
+ jiffies + msecs_to_jiffies(CMDQ_TIMEOUT_MS));
+ } else {
+ WARN_ON(cmdq_thread_suspend(cmdq, thread) < 0);
+ curr_pa = readl(thread->base + CMDQ_THR_CURR_ADDR);
+ end_pa = readl(thread->base + CMDQ_THR_END_ADDR);
+
+ /*
+ * Atomic execution should remove the following wfe, i.e. only
+ * wait event at first task, and prevent to pause when running.
+ */
+ if (thread->atomic_exec) {
+ /* GCE is executing if command is not WFE */
+ if (!cmdq_thread_is_in_wfe(thread)) {
+ cmdq_thread_resume(thread);
+ cmdq_thread_wait_end(thread, end_pa);
+ WARN_ON(cmdq_thread_suspend(cmdq, thread) < 0);
+ /* set to this task directly */
+ writel(task->pa_base,
+ thread->base + CMDQ_THR_CURR_ADDR);
+ } else {
+ cmdq_task_insert_into_thread(task);
+ cmdq_task_remove_wfe(task);
+ smp_mb(); /* modify jump before enable thread */
+ }
+ } else {
+ /* check boundary */
+ if (curr_pa == end_pa - CMDQ_INST_SIZE ||
+ curr_pa == end_pa) {
+ /* set to this task directly */
+ writel(task->pa_base,
+ thread->base + CMDQ_THR_CURR_ADDR);
+ } else {
+ cmdq_task_insert_into_thread(task);
+ smp_mb(); /* modify jump before enable thread */
+ }
+ }
+ writel(task->pa_base + pkt->cmd_buf_size,
+ thread->base + CMDQ_THR_END_ADDR);
+ cmdq_thread_resume(thread);
+ }
+ list_move_tail(&task->list_entry, &thread->task_busy_list);
+}
+
+static void cmdq_task_exec_done(struct cmdq_task *task, bool err)
+{
+ struct device *dev = task->cmdq->mbox.dev;
+ struct cmdq_cb_data cmdq_cb_data;
+
+ dma_unmap_single(dev, task->pa_base, task->pkt->cmd_buf_size,
+ DMA_TO_DEVICE);
+ if (task->pkt->cb.cb) {
+ cmdq_cb_data.err = err;
+ cmdq_cb_data.data = task->pkt->cb.data;
+ task->pkt->cb.cb(cmdq_cb_data);
+ }
+ list_del(&task->list_entry);
+}
+
+static void cmdq_task_handle_error(struct cmdq_task *task)
+{
+ struct cmdq_thread *thread = task->thread;
+ struct cmdq_task *next_task;
+
+ dev_err(task->cmdq->mbox.dev, "task 0x%p error\n", task);
+ WARN_ON(cmdq_thread_suspend(task->cmdq, thread) < 0);
+ next_task = list_first_entry_or_null(&thread->task_busy_list,
+ struct cmdq_task, list_entry);
+ if (next_task)
+ writel(next_task->pa_base, thread->base + CMDQ_THR_CURR_ADDR);
+ cmdq_thread_resume(thread);
+}
+
+static void cmdq_thread_irq_handler(struct cmdq *cmdq,
+ struct cmdq_thread *thread)
+{
+ struct cmdq_task *task, *tmp, *curr_task = NULL;
+ u32 curr_pa, irq_flag, task_end_pa;
+ bool err;
+
+ irq_flag = readl(thread->base + CMDQ_THR_IRQ_STATUS);
+ writel(~irq_flag, thread->base + CMDQ_THR_IRQ_STATUS);
+
+ /*
+ * When ISR call this function, another CPU core could run
+ * "release task" right before we acquire the spin lock, and thus
+ * reset / disable this GCE thread, so we need to check the enable
+ * bit of this GCE thread.
+ */
+ if (!(readl(thread->base + CMDQ_THR_ENABLE_TASK) & CMDQ_THR_ENABLED))
+ return;
+
+ if (irq_flag & CMDQ_THR_IRQ_ERROR)
+ err = true;
+ else if (irq_flag & CMDQ_THR_IRQ_DONE)
+ err = false;
+ else
+ return;
+
+ curr_pa = readl(thread->base + CMDQ_THR_CURR_ADDR);
+
+ list_for_each_entry_safe(task, tmp, &thread->task_busy_list,
+ list_entry) {
+ task_end_pa = task->pa_base + task->pkt->cmd_buf_size;
+ if (curr_pa >= task->pa_base && curr_pa < task_end_pa)
+ curr_task = task;
+
+ if (!curr_task || curr_pa == task_end_pa - CMDQ_INST_SIZE) {
+ cmdq_task_exec_done(task, false);
+ kfree(task);
+ } else if (err) {
+ cmdq_task_exec_done(task, true);
+ cmdq_task_handle_error(curr_task);
+ kfree(task);
+ }
+
+ if (curr_task)
+ break;
+ }
+
+ if (list_empty(&thread->task_busy_list)) {
+ cmdq_thread_disable(cmdq, thread);
+ clk_disable(cmdq->clock);
+ } else {
+ mod_timer(&thread->timeout,
+ jiffies + msecs_to_jiffies(CMDQ_TIMEOUT_MS));
+ }
+}
+
+static irqreturn_t cmdq_irq_handler(int irq, void *dev)
+{
+ struct cmdq *cmdq = dev;
+ unsigned long irq_status, flags = 0L;
+ int bit;
+
+ irq_status = readl(cmdq->base + CMDQ_CURR_IRQ_STATUS) & CMDQ_IRQ_MASK;
+ if (!(irq_status ^ CMDQ_IRQ_MASK))
+ return IRQ_NONE;
+
+ for_each_clear_bit(bit, &irq_status, fls(CMDQ_IRQ_MASK)) {
+ struct cmdq_thread *thread = &cmdq->thread[bit];
+
+ spin_lock_irqsave(&thread->chan->lock, flags);
+ cmdq_thread_irq_handler(cmdq, thread);
+ spin_unlock_irqrestore(&thread->chan->lock, flags);
+ }
+ return IRQ_HANDLED;
+}
+
+static void cmdq_thread_handle_timeout(unsigned long data)
+{
+ struct cmdq_thread *thread = (struct cmdq_thread *)data;
+ struct cmdq *cmdq = container_of(thread->chan->mbox, struct cmdq, mbox);
+ struct cmdq_task *task, *tmp;
+ unsigned long flags;
+
+ spin_lock_irqsave(&thread->chan->lock, flags);
+ WARN_ON(cmdq_thread_suspend(cmdq, thread) < 0);
+
+ /*
+ * Although IRQ is disabled, GCE continues to execute.
+ * It may have pending IRQ before GCE thread is suspended,
+ * so check this condition again.
+ */
+ cmdq_thread_irq_handler(cmdq, thread);
+
+ if (list_empty(&thread->task_busy_list)) {
+ cmdq_thread_resume(thread);
+ spin_unlock_irqrestore(&thread->chan->lock, flags);
+ return;
+ }
+
+ dev_err(cmdq->mbox.dev, "timeout\n");
+ list_for_each_entry_safe(task, tmp, &thread->task_busy_list,
+ list_entry) {
+ cmdq_task_exec_done(task, true);
+ kfree(task);
+ }
+
+ cmdq_thread_resume(thread);
+ cmdq_thread_disable(cmdq, thread);
+ clk_disable(cmdq->clock);
+ spin_unlock_irqrestore(&thread->chan->lock, flags);
+}
+
+static int cmdq_suspend(struct device *dev)
+{
+ struct cmdq *cmdq = dev_get_drvdata(dev);
+ struct cmdq_thread *thread;
+ int i;
+ bool task_running = false;
+
+ cmdq->suspended = true;
+
+ for (i = 0; i < ARRAY_SIZE(cmdq->thread); i++) {
+ thread = &cmdq->thread[i];
+ if (!list_empty(&thread->task_busy_list)) {
+ task_running = true;
+ break;
+ }
+ }
+
+ if (task_running)
+ dev_warn(dev, "exist running task(s) in suspend\n");
+
+ clk_unprepare(cmdq->clock);
+ return 0;
+}
+
+static int cmdq_resume(struct device *dev)
+{
+ struct cmdq *cmdq = dev_get_drvdata(dev);
+
+ WARN_ON(clk_prepare(cmdq->clock) < 0);
+ cmdq->suspended = false;
+ return 0;
+}
+
+static int cmdq_remove(struct platform_device *pdev)
+{
+ struct cmdq *cmdq = platform_get_drvdata(pdev);
+
+ mbox_controller_unregister(&cmdq->mbox);
+ clk_unprepare(cmdq->clock);
+ return 0;
+}
+
+static int cmdq_mbox_send_data(struct mbox_chan *chan, void *data)
+{
+ cmdq_task_exec(data, chan->con_priv);
+ return 0;
+}
+
+static int cmdq_mbox_startup(struct mbox_chan *chan)
+{
+ return 0;
+}
+
+static void cmdq_mbox_shutdown(struct mbox_chan *chan)
+{
+}
+
+static bool cmdq_mbox_last_tx_done(struct mbox_chan *chan)
+{
+ return true;
+}
+
+static const struct mbox_chan_ops cmdq_mbox_chan_ops = {
+ .send_data = cmdq_mbox_send_data,
+ .startup = cmdq_mbox_startup,
+ .shutdown = cmdq_mbox_shutdown,
+ .last_tx_done = cmdq_mbox_last_tx_done,
+};
+
+static struct mbox_chan *cmdq_xlate(struct mbox_controller *mbox,
+ const struct of_phandle_args *sp)
+{
+ int ind = sp->args[0];
+ struct cmdq_thread *thread;
+
+ if (ind >= mbox->num_chans)
+ return ERR_PTR(-EINVAL);
+
+ thread = mbox->chans[ind].con_priv;
+ thread->atomic_exec = (sp->args[1] != 0);
+ thread->chan = &mbox->chans[ind];
+
+ return &mbox->chans[ind];
+}
+
+static int cmdq_probe(struct platform_device *pdev)
+{
+ struct device *dev = &pdev->dev;
+ struct resource *res;
+ struct cmdq *cmdq;
+ int err, i;
+
+ cmdq = devm_kzalloc(dev, sizeof(*cmdq), GFP_KERNEL);
+ if (!cmdq)
+ return -ENOMEM;
+
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ cmdq->base = devm_ioremap_resource(dev, res);
+ if (IS_ERR(cmdq->base)) {
+ dev_err(dev, "failed to ioremap gce\n");
+ return PTR_ERR(cmdq->base);
+ }
+
+ cmdq->irq = platform_get_irq(pdev, 0);
+ if (!cmdq->irq) {
+ dev_err(dev, "failed to get irq\n");
+ return -EINVAL;
+ }
+ err = devm_request_irq(dev, cmdq->irq, cmdq_irq_handler, IRQF_SHARED,
+ "mtk_cmdq", cmdq);
+ if (err < 0) {
+ dev_err(dev, "failed to register ISR (%d)\n", err);
+ return err;
+ }
+
+ dev_dbg(dev, "cmdq device: addr:0x%p, va:0x%p, irq:%d\n",
+ dev, cmdq->base, cmdq->irq);
+
+ cmdq->clock = devm_clk_get(dev, "gce");
+ if (IS_ERR(cmdq->clock)) {
+ dev_err(dev, "failed to get gce clk\n");
+ return PTR_ERR(cmdq->clock);
+ }
+
+ cmdq->mbox.dev = dev;
+ cmdq->mbox.chans = devm_kcalloc(dev, CMDQ_THR_MAX_COUNT,
+ sizeof(*cmdq->mbox.chans), GFP_KERNEL);
+ if (!cmdq->mbox.chans)
+ return -ENOMEM;
+
+ cmdq->mbox.num_chans = CMDQ_THR_MAX_COUNT;
+ cmdq->mbox.ops = &cmdq_mbox_chan_ops;
+ cmdq->mbox.of_xlate = cmdq_xlate;
+
+ /* make use of TXDONE_BY_ACK */
+ cmdq->mbox.txdone_irq = false;
+ cmdq->mbox.txdone_poll = false;
+
+ for (i = 0; i < ARRAY_SIZE(cmdq->thread); i++) {
+ cmdq->thread[i].base = cmdq->base + CMDQ_THR_BASE +
+ CMDQ_THR_SIZE * i;
+ INIT_LIST_HEAD(&cmdq->thread[i].task_busy_list);
+ init_timer(&cmdq->thread[i].timeout);
+ cmdq->thread[i].timeout.function = cmdq_thread_handle_timeout;
+ cmdq->thread[i].timeout.data = (unsigned long)&cmdq->thread[i];
+ cmdq->mbox.chans[i].con_priv = &cmdq->thread[i];
+ }
+
+ err = mbox_controller_register(&cmdq->mbox);
+ if (err < 0) {
+ dev_err(dev, "failed to register mailbox: %d\n", err);
+ return err;
+ }
+
+ platform_set_drvdata(pdev, cmdq);
+ WARN_ON(clk_prepare(cmdq->clock) < 0);
+
+ return 0;
+}
+
+static const struct dev_pm_ops cmdq_pm_ops = {
+ .suspend = cmdq_suspend,
+ .resume = cmdq_resume,
+};
+
+static const struct of_device_id cmdq_of_ids[] = {
+ {.compatible = "mediatek,mt8173-gce",},
+ {}
+};
+
+static struct platform_driver cmdq_drv = {
+ .probe = cmdq_probe,
+ .remove = cmdq_remove,
+ .driver = {
+ .name = "mtk_cmdq",
+ .pm = &cmdq_pm_ops,
+ .of_match_table = cmdq_of_ids,
+ }
+};
+
+builtin_platform_driver(cmdq_drv);
diff --git a/include/linux/mailbox/mtk-cmdq-mailbox.h b/include/linux/mailbox/mtk-cmdq-mailbox.h
new file mode 100644
index 0000000..3433c64
--- /dev/null
+++ b/include/linux/mailbox/mtk-cmdq-mailbox.h
@@ -0,0 +1,75 @@
+/*
+ * Copyright (c) 2015 MediaTek Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef __MTK_CMDQ_MAILBOX_H__
+#define __MTK_CMDQ_MAILBOX_H__
+
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+
+#define CMDQ_INST_SIZE 8 /* instruction is 64-bit */
+#define CMDQ_OP_CODE_SHIFT 24
+#define CMDQ_JUMP_PASS CMDQ_INST_SIZE
+
+#define CMDQ_WFE_UPDATE BIT(31)
+#define CMDQ_WFE_WAIT BIT(15)
+#define CMDQ_WFE_WAIT_VALUE 0x1
+
+/*
+ * CMDQ_CODE_MASK:
+ * set write mask
+ * format: op mask
+ * CMDQ_CODE_WRITE:
+ * write value into target register
+ * format: op subsys address value
+ * CMDQ_CODE_JUMP:
+ * jump by offset
+ * format: op offset
+ * CMDQ_CODE_WFE:
+ * wait for event and clear
+ * it is just clear if no wait
+ * format: [wait] op event update:1 to_wait:1 wait:1
+ * [clear] op event update:1 to_wait:0 wait:0
+ * CMDQ_CODE_EOC:
+ * end of command
+ * format: op irq_flag
+ */
+enum cmdq_code {
+ CMDQ_CODE_MASK = 0x02,
+ CMDQ_CODE_WRITE = 0x04,
+ CMDQ_CODE_JUMP = 0x10,
+ CMDQ_CODE_WFE = 0x20,
+ CMDQ_CODE_EOC = 0x40,
+};
+
+struct cmdq_cb_data {
+ bool err;
+ void *data;
+};
+
+typedef void (*cmdq_async_flush_cb)(struct cmdq_cb_data data);
+
+struct cmdq_task_cb {
+ cmdq_async_flush_cb cb;
+ void *data;
+};
+
+struct cmdq_pkt {
+ void *va_base;
+ size_t cmd_buf_size; /* command occupied size */
+ size_t buf_size; /* real buffer size */
+ struct cmdq_task_cb cb;
+};
+
+#endif /* __MTK_CMDQ_MAILBOX_H__ */
--
1.9.1
^ permalink raw reply related
* [PATCH v20 3/4] arm64: dts: mt8173: Add GCE node
From: HS Liao @ 2017-01-04 3:06 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483499169-16329-1-git-send-email-hs.liao@mediatek.com>
This patch adds the device node of the GCE hardware for CMDQ module.
Signed-off-by: HS Liao <hs.liao@mediatek.com>
---
arch/arm64/boot/dts/mediatek/mt8173.dtsi | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/arch/arm64/boot/dts/mediatek/mt8173.dtsi b/arch/arm64/boot/dts/mediatek/mt8173.dtsi
index 12e7027..9f93447 100644
--- a/arch/arm64/boot/dts/mediatek/mt8173.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8173.dtsi
@@ -422,6 +422,16 @@
status = "disabled";
};
+ gce: gce at 10212000 {
+ compatible = "mediatek,mt8173-gce";
+ reg = <0 0x10212000 0 0x1000>;
+ interrupts = <GIC_SPI 135 IRQ_TYPE_LEVEL_LOW>;
+ clocks = <&infracfg CLK_INFRA_GCE>;
+ clock-names = "gce";
+
+ #mbox-cells = <2>;
+ };
+
mipi_tx0: mipi-dphy at 10215000 {
compatible = "mediatek,mt8173-mipi-tx";
reg = <0 0x10215000 0 0x1000>;
--
1.9.1
^ permalink raw reply related
* [PATCH v20 4/4] soc: mediatek: Add Mediatek CMDQ helper
From: HS Liao @ 2017-01-04 3:06 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483499169-16329-1-git-send-email-hs.liao@mediatek.com>
Add Mediatek CMDQ helper to create CMDQ packet and assemble GCE op code.
Signed-off-by: HS Liao <hs.liao@mediatek.com>
---
drivers/soc/mediatek/Kconfig | 12 ++
drivers/soc/mediatek/Makefile | 1 +
drivers/soc/mediatek/mtk-cmdq-helper.c | 310 +++++++++++++++++++++++++++++++++
include/linux/soc/mediatek/mtk-cmdq.h | 174 ++++++++++++++++++
4 files changed, 497 insertions(+)
create mode 100644 drivers/soc/mediatek/mtk-cmdq-helper.c
create mode 100644 include/linux/soc/mediatek/mtk-cmdq.h
diff --git a/drivers/soc/mediatek/Kconfig b/drivers/soc/mediatek/Kconfig
index 609bb34..2f145d8 100644
--- a/drivers/soc/mediatek/Kconfig
+++ b/drivers/soc/mediatek/Kconfig
@@ -1,6 +1,18 @@
#
# MediaTek SoC drivers
#
+config MTK_CMDQ
+ bool "MediaTek CMDQ Support"
+ depends on ARM64 && ( ARCH_MEDIATEK || COMPILE_TEST )
+ select MAILBOX
+ select MTK_CMDQ_MBOX
+ select MTK_INFRACFG
+ help
+ Say yes here to add support for the MediaTek Command Queue (CMDQ)
+ driver. The CMDQ is used to help read/write registers with critical
+ time limitation, such as updating display configuration during the
+ vblank.
+
config MTK_INFRACFG
bool "MediaTek INFRACFG Support"
depends on ARCH_MEDIATEK || COMPILE_TEST
diff --git a/drivers/soc/mediatek/Makefile b/drivers/soc/mediatek/Makefile
index 12998b0..64ce5ee 100644
--- a/drivers/soc/mediatek/Makefile
+++ b/drivers/soc/mediatek/Makefile
@@ -1,3 +1,4 @@
+obj-$(CONFIG_MTK_CMDQ) += mtk-cmdq-helper.o
obj-$(CONFIG_MTK_INFRACFG) += mtk-infracfg.o
obj-$(CONFIG_MTK_PMIC_WRAP) += mtk-pmic-wrap.o
obj-$(CONFIG_MTK_SCPSYS) += mtk-scpsys.o
diff --git a/drivers/soc/mediatek/mtk-cmdq-helper.c b/drivers/soc/mediatek/mtk-cmdq-helper.c
new file mode 100644
index 0000000..7809e65
--- /dev/null
+++ b/drivers/soc/mediatek/mtk-cmdq-helper.c
@@ -0,0 +1,310 @@
+/*
+ * Copyright (c) 2015 MediaTek Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/completion.h>
+#include <linux/errno.h>
+#include <linux/of_address.h>
+#include <linux/soc/mediatek/mtk-cmdq.h>
+
+#define CMDQ_SUBSYS_SHIFT 16
+#define CMDQ_ARG_A_WRITE_MASK 0xffff
+#define CMDQ_WRITE_ENABLE_MASK BIT(0)
+#define CMDQ_EOC_IRQ_EN BIT(0)
+#define CMDQ_EOC_CMD ((u64)((CMDQ_CODE_EOC << CMDQ_OP_CODE_SHIFT)) \
+ << 32 | CMDQ_EOC_IRQ_EN)
+
+struct cmdq_subsys {
+ u32 base;
+ int id;
+};
+
+static const struct cmdq_subsys gce_subsys[] = {
+ {0x1400, 1},
+ {0x1401, 2},
+ {0x1402, 3},
+};
+
+static int cmdq_subsys_base_to_id(u32 base)
+{
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(gce_subsys); i++)
+ if (gce_subsys[i].base == base)
+ return gce_subsys[i].id;
+ return -EFAULT;
+}
+
+static int cmdq_pkt_realloc_cmd_buffer(struct cmdq_pkt *pkt, size_t size)
+{
+ void *new_buf;
+
+ new_buf = krealloc(pkt->va_base, size, GFP_KERNEL | __GFP_ZERO);
+ if (!new_buf)
+ return -ENOMEM;
+ pkt->va_base = new_buf;
+ pkt->buf_size = size;
+ return 0;
+}
+
+struct cmdq_base *cmdq_register_device(struct device *dev)
+{
+ struct cmdq_base *cmdq_base;
+ struct resource res;
+ int subsys;
+ u32 base;
+
+ if (of_address_to_resource(dev->of_node, 0, &res))
+ return NULL;
+ base = (u32)res.start;
+
+ subsys = cmdq_subsys_base_to_id(base >> 16);
+ if (subsys < 0)
+ return NULL;
+
+ cmdq_base = devm_kmalloc(dev, sizeof(*cmdq_base), GFP_KERNEL);
+ if (!cmdq_base)
+ return NULL;
+ cmdq_base->subsys = subsys;
+ cmdq_base->base = base;
+
+ return cmdq_base;
+}
+EXPORT_SYMBOL(cmdq_register_device);
+
+struct cmdq_client *cmdq_mbox_create(struct device *dev, int index)
+{
+ struct cmdq_client *client;
+
+ client = kzalloc(sizeof(*client), GFP_KERNEL);
+ client->client.dev = dev;
+ client->client.tx_block = false;
+ client->chan = mbox_request_channel(&client->client, index);
+ return client;
+}
+EXPORT_SYMBOL(cmdq_mbox_create);
+
+void cmdq_mbox_destroy(struct cmdq_client *client)
+{
+ mbox_free_channel(client->chan);
+ kfree(client);
+}
+EXPORT_SYMBOL(cmdq_mbox_destroy);
+
+int cmdq_pkt_create(struct cmdq_pkt **pkt_ptr)
+{
+ struct cmdq_pkt *pkt;
+ int err;
+
+ pkt = kzalloc(sizeof(*pkt), GFP_KERNEL);
+ if (!pkt)
+ return -ENOMEM;
+ err = cmdq_pkt_realloc_cmd_buffer(pkt, PAGE_SIZE);
+ if (err < 0) {
+ kfree(pkt);
+ return err;
+ }
+ *pkt_ptr = pkt;
+ return 0;
+}
+EXPORT_SYMBOL(cmdq_pkt_create);
+
+void cmdq_pkt_destroy(struct cmdq_pkt *pkt)
+{
+ kfree(pkt->va_base);
+ kfree(pkt);
+}
+EXPORT_SYMBOL(cmdq_pkt_destroy);
+
+static bool cmdq_pkt_is_finalized(struct cmdq_pkt *pkt)
+{
+ u64 *expect_eoc;
+
+ if (pkt->cmd_buf_size < CMDQ_INST_SIZE << 1)
+ return false;
+
+ expect_eoc = pkt->va_base + pkt->cmd_buf_size - (CMDQ_INST_SIZE << 1);
+ if (*expect_eoc == CMDQ_EOC_CMD)
+ return true;
+
+ return false;
+}
+
+static int cmdq_pkt_append_command(struct cmdq_pkt *pkt, enum cmdq_code code,
+ u32 arg_a, u32 arg_b)
+{
+ u64 *cmd_ptr;
+ int err;
+
+ if (WARN_ON(cmdq_pkt_is_finalized(pkt)))
+ return -EBUSY;
+ if (unlikely(pkt->cmd_buf_size + CMDQ_INST_SIZE > pkt->buf_size)) {
+ err = cmdq_pkt_realloc_cmd_buffer(pkt, pkt->buf_size << 1);
+ if (err < 0)
+ return err;
+ }
+ cmd_ptr = pkt->va_base + pkt->cmd_buf_size;
+ (*cmd_ptr) = (u64)((code << CMDQ_OP_CODE_SHIFT) | arg_a) << 32 | arg_b;
+ pkt->cmd_buf_size += CMDQ_INST_SIZE;
+ return 0;
+}
+
+int cmdq_pkt_write(struct cmdq_pkt *pkt, u32 value, struct cmdq_base *base,
+ u32 offset)
+{
+ u32 arg_a = ((base->base + offset) & CMDQ_ARG_A_WRITE_MASK) |
+ (base->subsys << CMDQ_SUBSYS_SHIFT);
+ return cmdq_pkt_append_command(pkt, CMDQ_CODE_WRITE, arg_a, value);
+}
+EXPORT_SYMBOL(cmdq_pkt_write);
+
+int cmdq_pkt_write_mask(struct cmdq_pkt *pkt, u32 value,
+ struct cmdq_base *base, u32 offset, u32 mask)
+{
+ u32 offset_mask = offset;
+ int err;
+
+ if (mask != 0xffffffff) {
+ err = cmdq_pkt_append_command(pkt, CMDQ_CODE_MASK, 0, ~mask);
+ if (err < 0)
+ return err;
+ offset_mask |= CMDQ_WRITE_ENABLE_MASK;
+ }
+ return cmdq_pkt_write(pkt, value, base, offset_mask);
+}
+EXPORT_SYMBOL(cmdq_pkt_write_mask);
+
+static const u32 cmdq_event_value[CMDQ_MAX_EVENT] = {
+ /* Display start of frame(SOF) events */
+ [CMDQ_EVENT_DISP_OVL0_SOF] = 11,
+ [CMDQ_EVENT_DISP_OVL1_SOF] = 12,
+ [CMDQ_EVENT_DISP_RDMA0_SOF] = 13,
+ [CMDQ_EVENT_DISP_RDMA1_SOF] = 14,
+ [CMDQ_EVENT_DISP_RDMA2_SOF] = 15,
+ [CMDQ_EVENT_DISP_WDMA0_SOF] = 16,
+ [CMDQ_EVENT_DISP_WDMA1_SOF] = 17,
+ /* Display end of frame(EOF) events */
+ [CMDQ_EVENT_DISP_OVL0_EOF] = 39,
+ [CMDQ_EVENT_DISP_OVL1_EOF] = 40,
+ [CMDQ_EVENT_DISP_RDMA0_EOF] = 41,
+ [CMDQ_EVENT_DISP_RDMA1_EOF] = 42,
+ [CMDQ_EVENT_DISP_RDMA2_EOF] = 43,
+ [CMDQ_EVENT_DISP_WDMA0_EOF] = 44,
+ [CMDQ_EVENT_DISP_WDMA1_EOF] = 45,
+ /* Mutex end of frame(EOF) events */
+ [CMDQ_EVENT_MUTEX0_STREAM_EOF] = 53,
+ [CMDQ_EVENT_MUTEX1_STREAM_EOF] = 54,
+ [CMDQ_EVENT_MUTEX2_STREAM_EOF] = 55,
+ [CMDQ_EVENT_MUTEX3_STREAM_EOF] = 56,
+ [CMDQ_EVENT_MUTEX4_STREAM_EOF] = 57,
+ /* Display underrun events */
+ [CMDQ_EVENT_DISP_RDMA0_UNDERRUN] = 63,
+ [CMDQ_EVENT_DISP_RDMA1_UNDERRUN] = 64,
+ [CMDQ_EVENT_DISP_RDMA2_UNDERRUN] = 65,
+};
+
+int cmdq_pkt_wfe(struct cmdq_pkt *pkt, enum cmdq_event event)
+{
+ u32 arg_b;
+
+ if (event >= CMDQ_MAX_EVENT || event < 0)
+ return -EINVAL;
+
+ /*
+ * WFE arg_b
+ * bit 0-11: wait value
+ * bit 15: 1 - wait, 0 - no wait
+ * bit 16-27: update value
+ * bit 31: 1 - update, 0 - no update
+ */
+ arg_b = CMDQ_WFE_UPDATE | CMDQ_WFE_WAIT | CMDQ_WFE_WAIT_VALUE;
+ return cmdq_pkt_append_command(pkt, CMDQ_CODE_WFE,
+ cmdq_event_value[event], arg_b);
+}
+EXPORT_SYMBOL(cmdq_pkt_wfe);
+
+int cmdq_pkt_clear_event(struct cmdq_pkt *pkt, enum cmdq_event event)
+{
+ if (event >= CMDQ_MAX_EVENT || event < 0)
+ return -EINVAL;
+
+ return cmdq_pkt_append_command(pkt, CMDQ_CODE_WFE,
+ cmdq_event_value[event], CMDQ_WFE_UPDATE);
+}
+EXPORT_SYMBOL(cmdq_pkt_clear_event);
+
+static int cmdq_pkt_finalize(struct cmdq_pkt *pkt)
+{
+ int err;
+
+ if (cmdq_pkt_is_finalized(pkt))
+ return 0;
+
+ /* insert EOC and generate IRQ for each command iteration */
+ err = cmdq_pkt_append_command(pkt, CMDQ_CODE_EOC, 0, CMDQ_EOC_IRQ_EN);
+ if (err < 0)
+ return err;
+
+ /* JUMP to end */
+ err = cmdq_pkt_append_command(pkt, CMDQ_CODE_JUMP, 0, CMDQ_JUMP_PASS);
+ if (err < 0)
+ return err;
+
+ return 0;
+}
+
+int cmdq_pkt_flush_async(struct cmdq_client *client, struct cmdq_pkt *pkt,
+ cmdq_async_flush_cb cb, void *data)
+{
+ int err;
+
+ err = cmdq_pkt_finalize(pkt);
+ if (err < 0)
+ return err;
+
+ pkt->cb.cb = cb;
+ pkt->cb.data = data;
+
+ mbox_send_message(client->chan, pkt);
+ /* We can send next packet immediately, so just call txdone. */
+ mbox_client_txdone(client->chan, 0);
+
+ return 0;
+}
+EXPORT_SYMBOL(cmdq_pkt_flush_async);
+
+struct cmdq_flush_completion {
+ struct completion cmplt;
+ bool err;
+};
+
+static void cmdq_pkt_flush_cb(struct cmdq_cb_data data)
+{
+ struct cmdq_flush_completion *cmplt = data.data;
+
+ cmplt->err = data.err;
+ complete(&cmplt->cmplt);
+}
+
+int cmdq_pkt_flush(struct cmdq_client *client, struct cmdq_pkt *pkt)
+{
+ struct cmdq_flush_completion cmplt;
+ int err;
+
+ init_completion(&cmplt.cmplt);
+ err = cmdq_pkt_flush_async(client, pkt, cmdq_pkt_flush_cb, &cmplt);
+ if (err < 0)
+ return err;
+ wait_for_completion(&cmplt.cmplt);
+ return cmplt.err ? -EFAULT : 0;
+}
+EXPORT_SYMBOL(cmdq_pkt_flush);
diff --git a/include/linux/soc/mediatek/mtk-cmdq.h b/include/linux/soc/mediatek/mtk-cmdq.h
new file mode 100644
index 0000000..5b35d73
--- /dev/null
+++ b/include/linux/soc/mediatek/mtk-cmdq.h
@@ -0,0 +1,174 @@
+/*
+ * Copyright (c) 2015 MediaTek Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef __MTK_CMDQ_H__
+#define __MTK_CMDQ_H__
+
+#include <linux/mailbox_client.h>
+#include <linux/mailbox/mtk-cmdq-mailbox.h>
+
+/* display events in command queue(CMDQ) */
+enum cmdq_event {
+ /* Display start of frame(SOF) events */
+ CMDQ_EVENT_DISP_OVL0_SOF,
+ CMDQ_EVENT_DISP_OVL1_SOF,
+ CMDQ_EVENT_DISP_RDMA0_SOF,
+ CMDQ_EVENT_DISP_RDMA1_SOF,
+ CMDQ_EVENT_DISP_RDMA2_SOF,
+ CMDQ_EVENT_DISP_WDMA0_SOF,
+ CMDQ_EVENT_DISP_WDMA1_SOF,
+ /* Display end of frame(EOF) events */
+ CMDQ_EVENT_DISP_OVL0_EOF,
+ CMDQ_EVENT_DISP_OVL1_EOF,
+ CMDQ_EVENT_DISP_RDMA0_EOF,
+ CMDQ_EVENT_DISP_RDMA1_EOF,
+ CMDQ_EVENT_DISP_RDMA2_EOF,
+ CMDQ_EVENT_DISP_WDMA0_EOF,
+ CMDQ_EVENT_DISP_WDMA1_EOF,
+ /* Mutex end of frame(EOF) events */
+ CMDQ_EVENT_MUTEX0_STREAM_EOF,
+ CMDQ_EVENT_MUTEX1_STREAM_EOF,
+ CMDQ_EVENT_MUTEX2_STREAM_EOF,
+ CMDQ_EVENT_MUTEX3_STREAM_EOF,
+ CMDQ_EVENT_MUTEX4_STREAM_EOF,
+ /* Display underrun events */
+ CMDQ_EVENT_DISP_RDMA0_UNDERRUN,
+ CMDQ_EVENT_DISP_RDMA1_UNDERRUN,
+ CMDQ_EVENT_DISP_RDMA2_UNDERRUN,
+ /* Keep this at the end */
+ CMDQ_MAX_EVENT,
+};
+
+struct cmdq_pkt;
+
+struct cmdq_base {
+ int subsys;
+ u32 base;
+};
+
+struct cmdq_client {
+ struct mbox_client client;
+ struct mbox_chan *chan;
+};
+
+/**
+ * cmdq_register_device() - register device which needs CMDQ
+ * @dev: device for CMDQ to access its registers
+ *
+ * Return: cmdq_base pointer or NULL for failed
+ */
+struct cmdq_base *cmdq_register_device(struct device *dev);
+
+/**
+ * cmdq_mbox_create() - create CMDQ mailbox client and channel
+ * @dev: device of CMDQ mailbox client
+ * @index: index of CMDQ mailbox channel
+ *
+ * Return: CMDQ mailbox client pointer
+ */
+struct cmdq_client *cmdq_mbox_create(struct device *dev, int index);
+
+/**
+ * cmdq_mbox_destroy() - destroy CMDQ mailbox client and channel
+ * @client: the CMDQ mailbox client
+ */
+void cmdq_mbox_destroy(struct cmdq_client *client);
+
+/**
+ * cmdq_pkt_create() - create a CMDQ packet
+ * @pkt_ptr: CMDQ packet pointer to retrieve cmdq_pkt
+ *
+ * Return: 0 for success; else the error code is returned
+ */
+int cmdq_pkt_create(struct cmdq_pkt **pkt_ptr);
+
+/**
+ * cmdq_pkt_destroy() - destroy the CMDQ packet
+ * @pkt: the CMDQ packet
+ */
+void cmdq_pkt_destroy(struct cmdq_pkt *pkt);
+
+/**
+ * cmdq_pkt_write() - append write command to the CMDQ packet
+ * @pkt: the CMDQ packet
+ * @value: the specified target register value
+ * @base: the CMDQ base
+ * @offset: register offset from module base
+ *
+ * Return: 0 for success; else the error code is returned
+ */
+int cmdq_pkt_write(struct cmdq_pkt *pkt, u32 value,
+ struct cmdq_base *base, u32 offset);
+
+/**
+ * cmdq_pkt_write_mask() - append write command with mask to the CMDQ packet
+ * @pkt: the CMDQ packet
+ * @value: the specified target register value
+ * @base: the CMDQ base
+ * @offset: register offset from module base
+ * @mask: the specified target register mask
+ *
+ * Return: 0 for success; else the error code is returned
+ */
+int cmdq_pkt_write_mask(struct cmdq_pkt *pkt, u32 value,
+ struct cmdq_base *base, u32 offset, u32 mask);
+
+/**
+ * cmdq_pkt_wfe() - append wait for event command to the CMDQ packet
+ * @pkt: the CMDQ packet
+ * @event: the desired event type to "wait and CLEAR"
+ *
+ * Return: 0 for success; else the error code is returned
+ */
+int cmdq_pkt_wfe(struct cmdq_pkt *pkt, enum cmdq_event event);
+
+/**
+ * cmdq_pkt_clear_event() - append clear event command to the CMDQ packet
+ * @pkt: the CMDQ packet
+ * @event: the desired event to be cleared
+ *
+ * Return: 0 for success; else the error code is returned
+ */
+int cmdq_pkt_clear_event(struct cmdq_pkt *pkt, enum cmdq_event event);
+
+/**
+ * cmdq_pkt_flush() - trigger CMDQ to execute the CMDQ packet
+ * @client: the CMDQ mailbox client
+ * @pkt: the CMDQ packet
+ *
+ * Return: 0 for success; else the error code is returned
+ *
+ * Trigger CMDQ to execute the CMDQ packet. Note that this is a
+ * synchronous flush function. When the function returned, the recorded
+ * commands have been done.
+ */
+int cmdq_pkt_flush(struct cmdq_client *client, struct cmdq_pkt *pkt);
+
+/**
+ * cmdq_pkt_flush_async() - trigger CMDQ to asynchronously execute the CMDQ
+ * packet and call back at the end of done packet
+ * @client: the CMDQ mailbox client
+ * @pkt: the CMDQ packet
+ * @cb: called at the end of done packet
+ * @data: this data will pass back to cb
+ *
+ * Return: 0 for success; else the error code is returned
+ *
+ * Trigger CMDQ to asynchronously execute the CMDQ packet and call back
+ * at the end of done packet. Note that this is an ASYNC function. When the
+ * function returned, it may or may not be finished.
+ */
+int cmdq_pkt_flush_async(struct cmdq_client *client, struct cmdq_pkt *pkt,
+ cmdq_async_flush_cb cb, void *data);
+
+#endif /* __MTK_CMDQ_H__ */
--
1.9.1
^ permalink raw reply related
* [PATCH 1/4] watchdog: coh901327_wdt: Simplify error handling in probe function
From: Guenter Roeck @ 2017-01-04 3:25 UTC (permalink / raw)
To: linux-arm-kernel
Checking if there is no error followed by a goto if there is one is
confusing. Reverse the logic.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
drivers/watchdog/coh901327_wdt.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/watchdog/coh901327_wdt.c b/drivers/watchdog/coh901327_wdt.c
index a099b77fc0b9..dc97b2fd6c49 100644
--- a/drivers/watchdog/coh901327_wdt.c
+++ b/drivers/watchdog/coh901327_wdt.c
@@ -360,12 +360,10 @@ static int __init coh901327_probe(struct platform_device *pdev)
coh901327_wdt.parent = &pdev->dev;
ret = watchdog_register_device(&coh901327_wdt);
- if (ret == 0)
- dev_info(&pdev->dev,
- "initialized. timer margin=%d sec\n", margin);
- else
+ if (ret)
goto out_no_wdog;
+ dev_info(&pdev->dev, "initialized. timer margin=%d sec\n", margin);
return 0;
out_no_wdog:
--
2.7.4
^ permalink raw reply related
* [PATCH 2/4] watchdog: coh901327_wdt: Keep clock enabled after loading driver
From: Guenter Roeck @ 2017-01-04 3:25 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483500343-27113-1-git-send-email-linux@roeck-us.net>
Enabling the clock before accessing chip registers and disabling it
afterwards does not really make sense and only adds complexity to
the driver. In addition to that, a comment int the driver suggests
that it does not serve a useful purpose either.
"The watchdog block is of course always clocked, the
clk_enable()/clk_disable() calls are mainly for performing reference
counting higher up in the clock hierarchy."
Just keep the clock enabled instead.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
drivers/watchdog/coh901327_wdt.c | 21 +--------------------
1 file changed, 1 insertion(+), 20 deletions(-)
diff --git a/drivers/watchdog/coh901327_wdt.c b/drivers/watchdog/coh901327_wdt.c
index dc97b2fd6c49..1385a920df4f 100644
--- a/drivers/watchdog/coh901327_wdt.c
+++ b/drivers/watchdog/coh901327_wdt.c
@@ -74,11 +74,6 @@ static int irq;
static void __iomem *virtbase;
static struct device *parent;
-/*
- * The watchdog block is of course always clocked, the
- * clk_enable()/clk_disable() calls are mainly for performing reference
- * counting higher up in the clock hierarchy.
- */
static struct clk *clk;
/*
@@ -90,7 +85,6 @@ static void coh901327_enable(u16 timeout)
unsigned long freq;
unsigned long delay_ns;
- clk_enable(clk);
/* Restart timer if it is disabled */
val = readw(virtbase + U300_WDOG_D2R);
if (val == U300_WDOG_D2R_DISABLE_STATUS_DISABLED)
@@ -118,7 +112,6 @@ static void coh901327_enable(u16 timeout)
*/
(void) readw(virtbase + U300_WDOG_CR);
val = readw(virtbase + U300_WDOG_D2R);
- clk_disable(clk);
if (val != U300_WDOG_D2R_DISABLE_STATUS_ENABLED)
dev_err(parent,
"%s(): watchdog not enabled! D2R value %04x\n",
@@ -129,7 +122,6 @@ static void coh901327_disable(void)
{
u16 val;
- clk_enable(clk);
/* Disable the watchdog interrupt if it is active */
writew(0x0000U, virtbase + U300_WDOG_IMR);
/* If the watchdog is currently enabled, attempt to disable it */
@@ -144,7 +136,6 @@ static void coh901327_disable(void)
virtbase + U300_WDOG_D2R);
}
val = readw(virtbase + U300_WDOG_D2R);
- clk_disable(clk);
if (val != U300_WDOG_D2R_DISABLE_STATUS_DISABLED)
dev_err(parent,
"%s(): watchdog not disabled! D2R value %04x\n",
@@ -165,11 +156,9 @@ static int coh901327_stop(struct watchdog_device *wdt_dev)
static int coh901327_ping(struct watchdog_device *wdd)
{
- clk_enable(clk);
/* Feed the watchdog */
writew(U300_WDOG_FR_FEED_RESTART_TIMER,
virtbase + U300_WDOG_FR);
- clk_disable(clk);
return 0;
}
@@ -177,13 +166,11 @@ static int coh901327_settimeout(struct watchdog_device *wdt_dev,
unsigned int time)
{
wdt_dev->timeout = time;
- clk_enable(clk);
/* Set new timeout value */
writew(time * 100, virtbase + U300_WDOG_TR);
/* Feed the dog */
writew(U300_WDOG_FR_FEED_RESTART_TIMER,
virtbase + U300_WDOG_FR);
- clk_disable(clk);
return 0;
}
@@ -191,13 +178,11 @@ static unsigned int coh901327_gettimeleft(struct watchdog_device *wdt_dev)
{
u16 val;
- clk_enable(clk);
/* Read repeatedly until the value is stable! */
val = readw(virtbase + U300_WDOG_CR);
while (val & U300_WDOG_CR_VALID_IND)
val = readw(virtbase + U300_WDOG_CR);
val &= U300_WDOG_CR_COUNT_VALUE_MASK;
- clk_disable(clk);
if (val != 0)
val /= 100;
@@ -221,13 +206,11 @@ static irqreturn_t coh901327_interrupt(int irq, void *data)
* to prevent a watchdog reset by feeding the watchdog at this
* point.
*/
- clk_enable(clk);
val = readw(virtbase + U300_WDOG_IER);
if (val == U300_WDOG_IER_WILL_BARK_IRQ_EVENT_IND)
writew(U300_WDOG_IER_WILL_BARK_IRQ_ACK_ENABLE,
virtbase + U300_WDOG_IER);
writew(0x0000U, virtbase + U300_WDOG_IMR);
- clk_disable(clk);
dev_crit(parent, "watchdog is barking!\n");
return IRQ_HANDLED;
}
@@ -263,7 +246,7 @@ static int __exit coh901327_remove(struct platform_device *pdev)
watchdog_unregister_device(&coh901327_wdt);
coh901327_disable();
free_irq(irq, pdev);
- clk_unprepare(clk);
+ clk_disable_unprepare(clk);
clk_put(clk);
iounmap(virtbase);
release_mem_region(phybase, physize);
@@ -352,8 +335,6 @@ static int __init coh901327_probe(struct platform_device *pdev)
goto out_no_irq;
}
- clk_disable(clk);
-
ret = watchdog_init_timeout(&coh901327_wdt, margin, &pdev->dev);
if (ret < 0)
coh901327_wdt.timeout = 60;
--
2.7.4
^ permalink raw reply related
* [PATCH 3/4] watchdog: coh901327_wdt: Use devm_ioremap_resource to map resources
From: Guenter Roeck @ 2017-01-04 3:25 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483500343-27113-1-git-send-email-linux@roeck-us.net>
Map resources using devm_ioremap_resource() to simplify error handling.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
drivers/watchdog/coh901327_wdt.c | 31 +++++--------------------------
1 file changed, 5 insertions(+), 26 deletions(-)
diff --git a/drivers/watchdog/coh901327_wdt.c b/drivers/watchdog/coh901327_wdt.c
index 1385a920df4f..986222efe174 100644
--- a/drivers/watchdog/coh901327_wdt.c
+++ b/drivers/watchdog/coh901327_wdt.c
@@ -68,8 +68,6 @@
/* Default timeout in seconds = 1 minute */
static unsigned int margin = 60;
-static resource_size_t phybase;
-static resource_size_t physize;
static int irq;
static void __iomem *virtbase;
static struct device *parent;
@@ -248,8 +246,6 @@ static int __exit coh901327_remove(struct platform_device *pdev)
free_irq(irq, pdev);
clk_disable_unprepare(clk);
clk_put(clk);
- iounmap(virtbase);
- release_mem_region(phybase, physize);
return 0;
}
@@ -259,30 +255,18 @@ static int __init coh901327_probe(struct platform_device *pdev)
u16 val;
struct resource *res;
- res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
- if (!res)
- return -ENOENT;
-
parent = &pdev->dev;
- physize = resource_size(res);
- phybase = res->start;
-
- if (request_mem_region(phybase, physize, DRV_NAME) == NULL) {
- ret = -EBUSY;
- goto out;
- }
- virtbase = ioremap(phybase, physize);
- if (!virtbase) {
- ret = -ENOMEM;
- goto out_no_remap;
- }
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ virtbase = devm_ioremap_resource(&pdev->dev, res);
+ if (IS_ERR(virtbase))
+ return PTR_ERR(virtbase);
clk = clk_get(&pdev->dev, NULL);
if (IS_ERR(clk)) {
ret = PTR_ERR(clk);
dev_err(&pdev->dev, "could not get clock\n");
- goto out_no_clk;
+ return ret;
}
ret = clk_prepare_enable(clk);
if (ret) {
@@ -353,11 +337,6 @@ static int __init coh901327_probe(struct platform_device *pdev)
clk_disable_unprepare(clk);
out_no_clk_enable:
clk_put(clk);
-out_no_clk:
- iounmap(virtbase);
-out_no_remap:
- release_mem_region(phybase, SZ_4K);
-out:
return ret;
}
--
2.7.4
^ permalink raw reply related
* [PATCH 4/4] watchdog: coh901327_wdt: Use dev variable instead of pdev->dev
From: Guenter Roeck @ 2017-01-04 3:25 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483500343-27113-1-git-send-email-linux@roeck-us.net>
Use a local dev variable instead of dereferencing pdev->dev several
times in the probe function to make the code easier to read.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
drivers/watchdog/coh901327_wdt.c | 34 +++++++++++++++-------------------
1 file changed, 15 insertions(+), 19 deletions(-)
diff --git a/drivers/watchdog/coh901327_wdt.c b/drivers/watchdog/coh901327_wdt.c
index 986222efe174..38dd60f0cfcc 100644
--- a/drivers/watchdog/coh901327_wdt.c
+++ b/drivers/watchdog/coh901327_wdt.c
@@ -251,60 +251,56 @@ static int __exit coh901327_remove(struct platform_device *pdev)
static int __init coh901327_probe(struct platform_device *pdev)
{
+ struct device *dev = &pdev->dev;
int ret;
u16 val;
struct resource *res;
- parent = &pdev->dev;
+ parent = dev;
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
- virtbase = devm_ioremap_resource(&pdev->dev, res);
+ virtbase = devm_ioremap_resource(dev, res);
if (IS_ERR(virtbase))
return PTR_ERR(virtbase);
- clk = clk_get(&pdev->dev, NULL);
+ clk = clk_get(dev, NULL);
if (IS_ERR(clk)) {
ret = PTR_ERR(clk);
- dev_err(&pdev->dev, "could not get clock\n");
+ dev_err(dev, "could not get clock\n");
return ret;
}
ret = clk_prepare_enable(clk);
if (ret) {
- dev_err(&pdev->dev, "could not prepare and enable clock\n");
+ dev_err(dev, "could not prepare and enable clock\n");
goto out_no_clk_enable;
}
val = readw(virtbase + U300_WDOG_SR);
switch (val) {
case U300_WDOG_SR_STATUS_TIMED_OUT:
- dev_info(&pdev->dev,
- "watchdog timed out since last chip reset!\n");
+ dev_info(dev, "watchdog timed out since last chip reset!\n");
coh901327_wdt.bootstatus |= WDIOF_CARDRESET;
/* Status will be cleared below */
break;
case U300_WDOG_SR_STATUS_NORMAL:
- dev_info(&pdev->dev,
- "in normal status, no timeouts have occurred.\n");
+ dev_info(dev, "in normal status, no timeouts have occurred.\n");
break;
default:
- dev_info(&pdev->dev,
- "contains an illegal status code (%08x)\n", val);
+ dev_info(dev, "contains an illegal status code (%08x)\n", val);
break;
}
val = readw(virtbase + U300_WDOG_D2R);
switch (val) {
case U300_WDOG_D2R_DISABLE_STATUS_DISABLED:
- dev_info(&pdev->dev, "currently disabled.\n");
+ dev_info(dev, "currently disabled.\n");
break;
case U300_WDOG_D2R_DISABLE_STATUS_ENABLED:
- dev_info(&pdev->dev,
- "currently enabled! (disabling it now)\n");
+ dev_info(dev, "currently enabled! (disabling it now)\n");
coh901327_disable();
break;
default:
- dev_err(&pdev->dev,
- "contains an illegal enable/disable code (%08x)\n",
+ dev_err(dev, "contains an illegal enable/disable code (%08x)\n",
val);
break;
}
@@ -319,16 +315,16 @@ static int __init coh901327_probe(struct platform_device *pdev)
goto out_no_irq;
}
- ret = watchdog_init_timeout(&coh901327_wdt, margin, &pdev->dev);
+ ret = watchdog_init_timeout(&coh901327_wdt, margin, dev);
if (ret < 0)
coh901327_wdt.timeout = 60;
- coh901327_wdt.parent = &pdev->dev;
+ coh901327_wdt.parent = dev;
ret = watchdog_register_device(&coh901327_wdt);
if (ret)
goto out_no_wdog;
- dev_info(&pdev->dev, "initialized. timer margin=%d sec\n", margin);
+ dev_info(dev, "initialized. timer margin=%d sec\n", margin);
return 0;
out_no_wdog:
--
2.7.4
^ permalink raw reply related
* [PATCH v4 0/3] ARM: dts: imx6: Support Poslab Savageboard dual & quad
From: Milo Kim @ 2017-01-04 4:55 UTC (permalink / raw)
To: linux-arm-kernel
Poslab Savageboard is the i.MX6 SoC base, but BSP code from the vendor is
not mainline u-boot and kernel. Personal reason of using this board is
testing etnaviv user-space driver, so I re-write device tree files based on
mainline kernel for the first step.
This patchset includes common DT file, dual and quad board files.
Supported components are
- Display: HDMI and LVDS panel
- eMMC and SD card
- Ethernet
- Pinmux configuration
- SATA: only for Savageboard quad
- UART1 for debug console
- USB host
Missing features are
- Audio (WM8903)
- USB OTG
- PMIC WM8326: default settings are used so no issue to bring-up the system
- MIPI DSI and CSI
Patches are tested on the Savageboard quad but the dual version should work
because the only difference between dual and quad is SATA support.
More information in http://www.savageboard.org
v4:
Fix the license text and add the vendor prefix.
Use generic node name for the backlight panel.
Sort alphabetically for the pinctrl nodes.
Remove unnecessary pinmux of HDMI CEC.
v3:
Specify the dtbs for i.MX6 build.
v2:
Fix DT node for regulator, phy-reset-gpios and iomuxc node.
Milo Kim (3):
ARM: dts: imx6: Add Savageboard common file
ARM: dts: imx6: Support Savageboard dual
ARM: dts: imx6: Support Savageboard quad
.../devicetree/bindings/vendor-prefixes.txt | 1 +
arch/arm/boot/dts/Makefile | 2 +
arch/arm/boot/dts/imx6dl-savageboard.dts | 51 +++++
arch/arm/boot/dts/imx6q-savageboard.dts | 55 +++++
arch/arm/boot/dts/imx6qdl-savageboard.dtsi | 255 +++++++++++++++++++++
5 files changed, 364 insertions(+)
create mode 100644 arch/arm/boot/dts/imx6dl-savageboard.dts
create mode 100644 arch/arm/boot/dts/imx6q-savageboard.dts
create mode 100644 arch/arm/boot/dts/imx6qdl-savageboard.dtsi
--
2.11.0
^ permalink raw reply
* [PATCH v4 1/3] ARM: dts: imx6: Add Savageboard common file
From: Milo Kim @ 2017-01-04 4:55 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20170104045553.26576-1-woogyom.kim@gmail.com>
* Memory
memblock for DDR3 1GB
* Regulator
3.3V for panel and backlight.
* Display
Enable HDMI and LVDS panel. Savageboard supports AVIC TM097TDH02 panel
which is compatible with Hannstar HSD100PXN1, so reuse it.
* Clock
The commit d28be499c45e6 ("ARM: dts: imx6qdl-sabresd: Allow HDMI and
LVDS to work simultaneously") is applied to support LVDS and HDMI output
at the same time.
* Pinmux
Support eMMC, ethernet, gpio key for power button, I2C, PWM, SD card
and UART.
* Others
Enable ethernet, UART1 debug, USB host, USDHC3 for microSD card and
USDHC4 for built-in eMMC storage.
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Milo Kim <woogyom.kim@gmail.com>
---
arch/arm/boot/dts/imx6qdl-savageboard.dtsi | 255 +++++++++++++++++++++++++++++
1 file changed, 255 insertions(+)
create mode 100644 arch/arm/boot/dts/imx6qdl-savageboard.dtsi
diff --git a/arch/arm/boot/dts/imx6qdl-savageboard.dtsi b/arch/arm/boot/dts/imx6qdl-savageboard.dtsi
new file mode 100644
index 000000000000..a616e3c400d3
--- /dev/null
+++ b/arch/arm/boot/dts/imx6qdl-savageboard.dtsi
@@ -0,0 +1,255 @@
+/*
+ * Copyright (C) 2017 Milo Kim <woogyom.kim@gmail.com>
+ *
+ * This file is dual-licensed: you can use it either under the terms
+ * of the GPL or the X11 license, at your option. Note that this dual
+ * licensing only applies to this file, and not this project as a
+ * whole.
+ *
+ * a) This file is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * Or, alternatively,
+ *
+ * b) Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include <dt-bindings/gpio/gpio.h>
+#include <dt-bindings/input/input.h>
+
+/ {
+ chosen {
+ stdout-path = &uart1;
+ };
+
+ memory at 10000000 {
+ device_type = "memory";
+ reg = <0x10000000 0x40000000>;
+ };
+
+ gpio-keys {
+ compatible = "gpio-keys";
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_gpio_keys>;
+
+ power {
+ gpios = <&gpio3 7 GPIO_ACTIVE_LOW>;
+ label = "Power Button";
+ linux,code = <KEY_POWER>;
+ wakeup-source;
+ };
+ };
+
+ panel {
+ compatible = "avic,tm097tdh02", "hannstar,hsd100pxn1";
+ backlight = <&panel_bl>;
+ power-supply = <®_3p3v>;
+
+ port {
+ panel_in: endpoint {
+ remote-endpoint = <&lvds0_out>;
+ };
+ };
+ };
+
+ panel_bl: backlight {
+ compatible = "pwm-backlight";
+ brightness-levels = <0 4 8 16 32 64 128 255>;
+ default-brightness-level = <4>;
+ power-supply = <®_3p3v>;
+ pwms = <&pwm1 0 10000>;
+ };
+
+ reg_3p3v: regulator-3p3v {
+ compatible = "regulator-fixed";
+ regulator-name = "3P3V";
+ regulator-min-microvolt = <3300000>;
+ regulator-max-microvolt = <3300000>;
+ regulator-always-on;
+ };
+};
+
+&clks {
+ assigned-clocks = <&clks IMX6QDL_CLK_LDB_DI0_SEL>,
+ <&clks IMX6QDL_CLK_LDB_DI1_SEL>;
+ assigned-clock-parents = <&clks IMX6QDL_CLK_PLL3_USB_OTG>,
+ <&clks IMX6QDL_CLK_PLL3_USB_OTG>;
+};
+
+&fec {
+ phy-mode = "rgmii";
+ phy-reset-gpios = <&gpio1 25 GPIO_ACTIVE_LOW>;
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_enet>;
+ status = "okay";
+};
+
+&hdmi {
+ ddc-i2c-bus = <&i2c2>;
+ status = "okay";
+};
+
+&i2c2 {
+ clock-frequency = <100000>;
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_i2c2>;
+ status = "okay";
+};
+
+&ldb {
+ status = "okay";
+
+ lvds-channel at 0 {
+ reg = <0>;
+ status = "okay";
+
+ port at 4 {
+ reg = <4>;
+
+ lvds0_out: endpoint {
+ remote-endpoint = <&panel_in>;
+ };
+ };
+ };
+};
+
+&pwm1 {
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_pwm1>;
+ status = "okay";
+};
+
+&uart1 {
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_uart1>;
+ status = "okay";
+};
+
+&usbh1 {
+ status = "okay";
+};
+
+/* SD card */
+&usdhc3 {
+ bus-width = <4>;
+ cd-gpios = <&gpio2 0 GPIO_ACTIVE_LOW>;
+ no-1-8-v;
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_sd>;
+ status = "okay";
+};
+
+/* eMMC */
+&usdhc4 {
+ bus-width = <8>;
+ keep-power-in-suspend;
+ no-1-8-v;
+ non-removable;
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_emmc>;
+ status = "okay";
+};
+
+&iomuxc {
+ pinctrl_emmc: emmcgrp {
+ fsl,pins = <
+ MX6QDL_PAD_SD4_CMD__SD4_CMD 0x17059
+ MX6QDL_PAD_SD4_CLK__SD4_CLK 0x10059
+ MX6QDL_PAD_SD4_DAT0__SD4_DATA0 0x17059
+ MX6QDL_PAD_SD4_DAT1__SD4_DATA1 0x17059
+ MX6QDL_PAD_SD4_DAT2__SD4_DATA2 0x17059
+ MX6QDL_PAD_SD4_DAT3__SD4_DATA3 0x17059
+ MX6QDL_PAD_SD4_DAT4__SD4_DATA4 0x17059
+ MX6QDL_PAD_SD4_DAT5__SD4_DATA5 0x17059
+ MX6QDL_PAD_SD4_DAT6__SD4_DATA6 0x17059
+ MX6QDL_PAD_SD4_DAT7__SD4_DATA7 0x17059
+ >;
+ };
+
+ pinctrl_enet: enetgrp {
+ fsl,pins = <
+ MX6QDL_PAD_ENET_MDIO__ENET_MDIO 0x1b0b0
+ MX6QDL_PAD_ENET_MDC__ENET_MDC 0x1b0b0
+ MX6QDL_PAD_RGMII_TXC__RGMII_TXC 0x1b030
+ MX6QDL_PAD_RGMII_TD0__RGMII_TD0 0x1b030
+ MX6QDL_PAD_RGMII_TD1__RGMII_TD1 0x1b030
+ MX6QDL_PAD_RGMII_TD2__RGMII_TD2 0x1b030
+ MX6QDL_PAD_RGMII_TD3__RGMII_TD3 0x1b030
+ MX6QDL_PAD_RGMII_TX_CTL__RGMII_TX_CTL 0x1b030
+ MX6QDL_PAD_ENET_REF_CLK__ENET_TX_CLK 0x1b0b0
+ MX6QDL_PAD_RGMII_RXC__RGMII_RXC 0x1b030
+ MX6QDL_PAD_RGMII_RD0__RGMII_RD0 0x1b030
+ MX6QDL_PAD_RGMII_RD1__RGMII_RD1 0x1b030
+ MX6QDL_PAD_RGMII_RD2__RGMII_RD2 0x1b030
+ MX6QDL_PAD_RGMII_RD3__RGMII_RD3 0x1b030
+ MX6QDL_PAD_RGMII_RX_CTL__RGMII_RX_CTL 0x1b030
+ /* PHY reset */
+ MX6QDL_PAD_ENET_CRS_DV__GPIO1_IO25 0x1b0b0
+ >;
+ };
+
+ pinctrl_gpio_keys: gpiokeysgrp {
+ fsl,pins = <
+ MX6QDL_PAD_EIM_DA7__GPIO3_IO07 0x1b0b1
+ >;
+ };
+
+ pinctrl_i2c2: i2c2grp {
+ fsl,pins = <
+ MX6QDL_PAD_KEY_COL3__I2C2_SCL 0x4001b8b1
+ MX6QDL_PAD_KEY_ROW3__I2C2_SDA 0x4001b8b1
+ >;
+ };
+
+ pinctrl_pwm1: pwm1grp {
+ fsl,pins = <
+ MX6QDL_PAD_SD1_DAT3__PWM1_OUT 0x1b0b1
+ >;
+ };
+
+ pinctrl_sd: sdgrp {
+ fsl,pins = <
+ MX6QDL_PAD_SD3_CMD__SD3_CMD 0x17059
+ MX6QDL_PAD_SD3_CLK__SD3_CLK 0x10059
+ MX6QDL_PAD_SD3_DAT0__SD3_DATA0 0x17059
+ MX6QDL_PAD_SD3_DAT1__SD3_DATA1 0x17059
+ MX6QDL_PAD_SD3_DAT2__SD3_DATA2 0x17059
+ MX6QDL_PAD_SD3_DAT3__SD3_DATA3 0x17059
+ /* CD pin */
+ MX6QDL_PAD_NANDF_D0__GPIO2_IO00 0x1b0b1
+ >;
+ };
+
+ pinctrl_uart1: uart1grp {
+ fsl,pins = <
+ MX6QDL_PAD_CSI0_DAT10__UART1_TX_DATA 0x1b0b1
+ MX6QDL_PAD_CSI0_DAT11__UART1_RX_DATA 0x1b0b1
+ >;
+ };
+};
--
2.11.0
^ permalink raw reply related
* [PATCH v4 2/3] ARM: dts: imx6: Support Savageboard dual
From: Milo Kim @ 2017-01-04 4:55 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20170104045553.26576-1-woogyom.kim@gmail.com>
Common savageboard DT file is used for board support.
Add the vendor name and specify the dtb file for i.MX6Q build.
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Milo Kim <woogyom.kim@gmail.com>
---
.../devicetree/bindings/vendor-prefixes.txt | 1 +
arch/arm/boot/dts/Makefile | 1 +
arch/arm/boot/dts/imx6dl-savageboard.dts | 51 ++++++++++++++++++++++
3 files changed, 53 insertions(+)
create mode 100644 arch/arm/boot/dts/imx6dl-savageboard.dts
diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt
index 16d3b5e7f5d1..88c33d827e51 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.txt
+++ b/Documentation/devicetree/bindings/vendor-prefixes.txt
@@ -227,6 +227,7 @@ pine64 Pine64
pixcir PIXCIR MICROELECTRONICS Co., Ltd
plathome Plat'Home Co., Ltd.
plda PLDA
+poslab Poslab Technology Co., Ltd.
powervr PowerVR (deprecated, use img)
pulsedlight PulsedLight, Inc
qca Qualcomm Atheros, Inc.
diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile
index cccdbcb557b6..fb7f904a5235 100644
--- a/arch/arm/boot/dts/Makefile
+++ b/arch/arm/boot/dts/Makefile
@@ -357,6 +357,7 @@ dtb-$(CONFIG_SOC_IMX6Q) += \
imx6dl-sabreauto.dtb \
imx6dl-sabrelite.dtb \
imx6dl-sabresd.dtb \
+ imx6dl-savageboard.dtb \
imx6dl-ts4900.dtb \
imx6dl-tx6dl-comtft.dtb \
imx6dl-tx6s-8034.dtb \
diff --git a/arch/arm/boot/dts/imx6dl-savageboard.dts b/arch/arm/boot/dts/imx6dl-savageboard.dts
new file mode 100644
index 000000000000..b95469c520a4
--- /dev/null
+++ b/arch/arm/boot/dts/imx6dl-savageboard.dts
@@ -0,0 +1,51 @@
+/*
+ * Copyright (C) 2017 Milo Kim <woogyom.kim@gmail.com>
+ *
+ * This file is dual-licensed: you can use it either under the terms
+ * of the GPL or the X11 license, at your option. Note that this dual
+ * licensing only applies to this file, and not this project as a
+ * whole.
+ *
+ * a) This file is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * Or, alternatively,
+ *
+ * b) Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+/dts-v1/;
+
+#include "imx6dl.dtsi"
+#include "imx6qdl-savageboard.dtsi"
+
+/ {
+ model = "Poslab SavageBoard Dual";
+ compatible = "poslab,imx6dl-savageboard", "fsl,imx6dl";
+};
--
2.11.0
^ permalink raw reply related
* [PATCH v4 3/3] ARM: dts: imx6: Support Savageboard quad
From: Milo Kim @ 2017-01-04 4:55 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20170104045553.26576-1-woogyom.kim@gmail.com>
Use common board file and support SATA interface additionally.
Specify the dtb file for i.MX6 build.
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Milo Kim <woogyom.kim@gmail.com>
---
arch/arm/boot/dts/Makefile | 1 +
arch/arm/boot/dts/imx6q-savageboard.dts | 55 +++++++++++++++++++++++++++++++++
2 files changed, 56 insertions(+)
create mode 100644 arch/arm/boot/dts/imx6q-savageboard.dts
diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile
index fb7f904a5235..2f21e59adc1e 100644
--- a/arch/arm/boot/dts/Makefile
+++ b/arch/arm/boot/dts/Makefile
@@ -403,6 +403,7 @@ dtb-$(CONFIG_SOC_IMX6Q) += \
imx6q-sabreauto.dtb \
imx6q-sabrelite.dtb \
imx6q-sabresd.dtb \
+ imx6q-savageboard.dtb \
imx6q-sbc6x.dtb \
imx6q-tbs2910.dtb \
imx6q-ts4900.dtb \
diff --git a/arch/arm/boot/dts/imx6q-savageboard.dts b/arch/arm/boot/dts/imx6q-savageboard.dts
new file mode 100644
index 000000000000..717ac62fc2cf
--- /dev/null
+++ b/arch/arm/boot/dts/imx6q-savageboard.dts
@@ -0,0 +1,55 @@
+/*
+ * Copyright (C) 2017 Milo Kim <woogyom.kim@gmail.com>
+ *
+ * This file is dual-licensed: you can use it either under the terms
+ * of the GPL or the X11 license, at your option. Note that this dual
+ * licensing only applies to this file, and not this project as a
+ * whole.
+ *
+ * a) This file is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This file is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * Or, alternatively,
+ *
+ * b) Permission is hereby granted, free of charge, to any person
+ * obtaining a copy of this software and associated documentation
+ * files (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following
+ * conditions:
+ *
+ * The above copyright notice and this permission notice shall be
+ * included in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+ * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+/dts-v1/;
+
+#include "imx6q.dtsi"
+#include "imx6qdl-savageboard.dtsi"
+
+/ {
+ model = "Poslab SavageBoard Quad";
+ compatible = "poslab,imx6q-savageboard", "fsl,imx6q";
+};
+
+&sata {
+ status = "okay";
+};
--
2.11.0
^ permalink raw reply related
* [PATCH] ARM: dts: am335x-icev2: Remove the duplicated pinmux setting
From: Lokesh Vutla @ 2017-01-04 5:09 UTC (permalink / raw)
To: linux-arm-kernel
There is no mmc sd card detect on am335x-ice board. But the spi0_cs1
pin being configured as mmcsd_cd. Removing it fixes the below warning
during boot.
[1.401303] pinctrl-single 44e10800.pinmux: pin 44e10960.0 already requested by 48030000.spi; cannot claim for 48060000.mmc
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
---
Logs:
Before this patch: http://pastebin.ubuntu.com/23737509/
After this patch: http://pastebin.ubuntu.com/23737467/
arch/arm/boot/dts/am335x-icev2.dts | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/arm/boot/dts/am335x-icev2.dts b/arch/arm/boot/dts/am335x-icev2.dts
index 1463df3b5b19..8ed46f9d79b7 100644
--- a/arch/arm/boot/dts/am335x-icev2.dts
+++ b/arch/arm/boot/dts/am335x-icev2.dts
@@ -170,7 +170,6 @@
AM33XX_IOPAD(0x8fc, PIN_INPUT_PULLUP | MUX_MODE0) /* (G16) mmc0_dat0.mmc0_dat0 */
AM33XX_IOPAD(0x900, PIN_INPUT_PULLUP | MUX_MODE0) /* (G17) mmc0_clk.mmc0_clk */
AM33XX_IOPAD(0x904, PIN_INPUT_PULLUP | MUX_MODE0) /* (G18) mmc0_cmd.mmc0_cmd */
- AM33XX_IOPAD(0x960, PIN_INPUT_PULLUP | MUX_MODE5) /* (C15) spi0_cs1.mmc0_sdcd */
>;
};
--
2.11.0
^ permalink raw reply related
* [PATCH v4 0/4] Add runtime PM support for clocks (on Exynos SoC example)
From: Chanwoo Choi @ 2017-01-04 5:25 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483104816-20885-1-git-send-email-m.szyprowski@samsung.com>
Dear Marek,
I like these patches because as you already said, the some clocks are included in the specific power domain such as exynos5433. I want to test these patches on Exynos5433-TM2 board. But, this series don't include the Device-Tree binding patches. Could you give the additional patches for test?
Best Regards,
Chanwoo Choi
On 2016? 12? 30? 22:33, Marek Szyprowski wrote:
> Dear All,
>
> This patchset adds runtime PM support to common clock framework. This is an
> attempt to implement support for clock controllers, which belongs to a power
> domain. This approach works surprisingly well on Exynos 5433 SoC, what allowed
> us to solve various freeze/crash issues related to power management.
>
> The main idea behind this patchset is to keep clock's controller power domain
> enabled every time when at least one of its clock is enabled or access to its
> registers is being made. Clock controller driver (clock provider) can
> supply a struct device pointer, which is the used by clock core for tracking and
> managing clock's controller runtime pm state. Each clk_prepare() operation will
> first call pm_runtime_get_sync() on the supplied device, while clk_unprepare()
> will do pm_runtime_put() at the end.
>
> This runtime PM feature has been tested with Exynos4412 (not included in this
> patchset) and Exynos5433 clocks drivers. Both have some clocks, which belongs to
> respective power domains and need special handling during power on/off
> procedures. Till now it wasn't handled at all, what caused various problems.
>
> Patches for clocks drivers change the way the clock provider is initialized.
> Instead of CLK_OF_DECLARE based initialization, a complete platform device driver
> infrastructure is being used. This is needed to let driver to use runtime PM
> feature and integrate with generic power domains. The side-effect of this change
> is a delay in clock provider registeration during system boot, so early
> initialized drivers might get EPROBEDEFER error when requesting their clocks.
> This is an issue for IOMMU drivers, so this patchset will be fully functional
> once the deferred probe for IOMMU will be merged.
>
> Patches are based on vanilla v4.10-rc1 kernel.
>
> Best regards
> Marek Szyprowski
> Samsung R&D Institute Poland
>
> Changelog:
>
> v4:
> - Removed patch for Exynos4412 clocks from the patchset. DT bindings for power
> domain for ISP sub-controller needs more discussion. It will be handled
> separately when the runtime PM for clocks feature gets merged.
> - Added patch with runtime PM support for Exynos AudioSS clock controller driver
> (needed to enable audio power domain on Exynos5 series).
> - Fixes in Exynos5433 driver:
> 1. added missing clock for Audio CMU
> 2. added support for FSYS CMU
> 3. improved support for DISP CMU (thanks to Andrzej Hajda for
> investigating that).
> - Rebased onto v4.10-rc1
> - Waiting for a review...
>
> v3: http://www.spinics.net/lists/arm-kernel/msg538122.html
> - Removed CLK_RUNTIME_PM flag, core now simply checks if runtime pm is enabled
> for the provided device during clock registration as suggested by Ulf
> - Simplified code for exynos4412 isp clock driver registration
> - Resolved some other minor issues pointed by Ulf clk core code
> - Rebased onto v4.9-rc1 and new version of IOMMU deferred probe patchset
>
> v2: https://www.spinics.net/lists/arm-kernel/msg532798.html
> - Simplified clk_pm_runtime_get/put functions, removed workaround for devices
> with disabled runtime pm. Such workaround is no longer needed since commit
> 4d23a5e84806b202d9231929c9507ef7cf7a0185 ("PM / Domains: Allow runtime PM
> during system PM phases").
> - Added CLK_RUNTIME_PM flag to indicate clocks, for which clock core should
> call runtime pm functions. This solves problem with clocks, for which struct
> device is already registered, but no runtime pm is enabled.
> - Extended commit messages according to Ulf suggestions.
> - Fixed some style issues pointed by Barlomiej.
>
> v1: http://www.spinics.net/lists/arm-kernel/msg528128.html
> - initial version
>
>
> Marek Szyprowski (4):
> clk: Add support for runtime PM
> clk: samsung: Add support for runtime PM
> clk: samsung: exynos5433: Add runtime PM support
> clk: samsung: exynos-audss: Use runtime PM
>
> .../devicetree/bindings/clock/clk-exynos-audss.txt | 6 +
> .../devicetree/bindings/clock/exynos5433-clock.txt | 16 +
> drivers/clk/clk.c | 111 +++++-
> drivers/clk/samsung/clk-exynos-audss.c | 62 +--
> drivers/clk/samsung/clk-exynos5433.c | 415 ++++++++++++++++-----
> drivers/clk/samsung/clk-pll.c | 2 +-
> drivers/clk/samsung/clk.c | 12 +-
> drivers/clk/samsung/clk.h | 7 +
> 8 files changed, 504 insertions(+), 127 deletions(-)
>
^ permalink raw reply
* [PATCH v5 1/4] soc: zte: Add header for PM domains specifiers
From: Shawn Guo @ 2017-01-04 5:29 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483489157-10782-1-git-send-email-baoyou.xie@linaro.org>
On Wed, Jan 04, 2017 at 08:19:14AM +0800, Baoyou Xie wrote:
> This patch adds header with values used for ZTE 2967
> SoC's power domain driver.
>
> Signed-off-by: Baoyou Xie <baoyou.xie@linaro.org>
> Reviewed-by: Shawn Guo <shawn.guo@linaro.org>
The tags like Reviewed-by, Acked-by etc, can only be added into your
patch after people give it explicitly. One of the examples is that Jun
Nie gave his Reviewed-by tag on your "[PATCH v2 2/2] soc: zte:
pm_domains: Add support for zx296718 board".
https://www.spinics.net/lists/arm-kernel/msg547282.html
I did not give my Reviewed-by tag on list, so you shouldn't add it here.
Actually, in case of this patch series, I will probably apply them
through ZTE -> arm-soc tree, when I feel it's ready. So I'm do not even
need to give the Reviewed-by tag.
Shawn
> ---
> include/dt-bindings/soc/zte,pm_domains.h | 23 +++++++++++++++++++++++
> 1 file changed, 23 insertions(+)
> create mode 100644 include/dt-bindings/soc/zte,pm_domains.h
>
> diff --git a/include/dt-bindings/soc/zte,pm_domains.h b/include/dt-bindings/soc/zte,pm_domains.h
> new file mode 100644
> index 0000000..01e9abc
> --- /dev/null
> +++ b/include/dt-bindings/soc/zte,pm_domains.h
> @@ -0,0 +1,23 @@
> +/*
> + * Copyright (C) 2017 Linaro Ltd.
> + *
> + * Author: Baoyou Xie <baoyou.xie@linaro.org>
> + * License terms: GNU General Public License (GPL) version 2
> + */
> +#ifndef _DT_BINDINGS_SOC_ZTE_PM_DOMAINS_H
> +#define _DT_BINDINGS_SOC_ZTE_PM_DOMAINS_H
> +
> +#define DM_ZX296718_SAPPU 0
> +#define DM_ZX296718_VDE 1 /* g1v6 */
> +#define DM_ZX296718_VCE 2 /* h1v6 */
> +#define DM_ZX296718_HDE 3 /* g2v2 */
> +#define DM_ZX296718_VIU 4
> +#define DM_ZX296718_USB20 5
> +#define DM_ZX296718_USB21 6
> +#define DM_ZX296718_USB30 7
> +#define DM_ZX296718_HSIC 8
> +#define DM_ZX296718_GMAC 9
> +#define DM_ZX296718_TS 10
> +#define DM_ZX296718_VOU 11
> +
> +#endif /* _DT_BINDINGS_SOC_ZTE_PM_DOMAINS_H */
> --
> 2.7.4
>
^ permalink raw reply
* [PATCH v3 0/2] Broadcom FlexRM ring manager support
From: Anup Patel @ 2017-01-04 5:34 UTC (permalink / raw)
To: linux-arm-kernel
The Broadcom FlexRM ring manager provides producer-consumer style
ring interface for offload engines on Broadcom iProc SoCs. We can
have one or more instances of Broadcom FlexRM ring manager in a SoC.
This patchset adds a mailbox driver for Broadcom FlexRM ring manager
which can be used by offload engine drivers as mailbox clients.
The Broadcom FlexRM mailbox driver is feature complete for RAID and
Crypto offload engines. We will have incremental patches in-future
for ring-level statistics using debugfs and minor optimizations.
This patchset is based on Linux-4.10-rc2 and it is also available
at flexrm-v3 branch of https://github.com/Broadcom/arm64-linux.git
Changes since v2:
- Rebased patches for Linux-4.10-rc2
Changes since v1:
- Use compatile string as brcm,iproc-flexrm-mbox
- Rephrase commit message and text in DT bindings patch
Anup Patel (2):
mailbox: Add driver for Broadcom FlexRM ring manager
dt-bindings: Add DT bindings info for FlexRM ring manager
.../bindings/mailbox/brcm,iproc-flexrm-mbox.txt | 60 ++
drivers/mailbox/Kconfig | 11 +
drivers/mailbox/Makefile | 2 +
drivers/mailbox/mailbox-flexrm/Makefile | 6 +
drivers/mailbox/mailbox-flexrm/flexrm-desc.c | 764 +++++++++++++++++++
drivers/mailbox/mailbox-flexrm/flexrm-desc.h | 47 ++
drivers/mailbox/mailbox-flexrm/flexrm-main.c | 829 +++++++++++++++++++++
include/linux/mailbox/brcm-message.h | 14 +-
8 files changed, 1729 insertions(+), 4 deletions(-)
create mode 100644 Documentation/devicetree/bindings/mailbox/brcm,iproc-flexrm-mbox.txt
create mode 100644 drivers/mailbox/mailbox-flexrm/Makefile
create mode 100644 drivers/mailbox/mailbox-flexrm/flexrm-desc.c
create mode 100644 drivers/mailbox/mailbox-flexrm/flexrm-desc.h
create mode 100644 drivers/mailbox/mailbox-flexrm/flexrm-main.c
--
2.7.4
^ permalink raw reply
* [PATCH v3 1/2] mailbox: Add driver for Broadcom FlexRM ring manager
From: Anup Patel @ 2017-01-04 5:34 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483508082-7008-1-git-send-email-anup.patel@broadcom.com>
Some of the Broadcom iProc SoCs have FlexRM ring manager
which provides a ring-based programming interface to various
offload engines (e.g. RAID, Crypto, etc).
This patch adds a common mailbox driver for Broadcom FlexRM
ring manager which can be shared by various offload engine
drivers (implemented as mailbox clients).
Reviewed-by: Ray Jui <ray.jui@broadcom.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
Reviewed-by: Pramod KUMAR <pramod.kumar@broadcom.com>
Signed-off-by: Anup Patel <anup.patel@broadcom.com>
---
drivers/mailbox/Kconfig | 11 +
drivers/mailbox/Makefile | 2 +
drivers/mailbox/mailbox-flexrm/Makefile | 6 +
drivers/mailbox/mailbox-flexrm/flexrm-desc.c | 764 ++++++++++++++++++++++++
drivers/mailbox/mailbox-flexrm/flexrm-desc.h | 47 ++
drivers/mailbox/mailbox-flexrm/flexrm-main.c | 829 +++++++++++++++++++++++++++
include/linux/mailbox/brcm-message.h | 14 +-
7 files changed, 1669 insertions(+), 4 deletions(-)
create mode 100644 drivers/mailbox/mailbox-flexrm/Makefile
create mode 100644 drivers/mailbox/mailbox-flexrm/flexrm-desc.c
create mode 100644 drivers/mailbox/mailbox-flexrm/flexrm-desc.h
create mode 100644 drivers/mailbox/mailbox-flexrm/flexrm-main.c
diff --git a/drivers/mailbox/Kconfig b/drivers/mailbox/Kconfig
index ceff415..305018c 100644
--- a/drivers/mailbox/Kconfig
+++ b/drivers/mailbox/Kconfig
@@ -152,4 +152,15 @@ config BCM_PDC_MBOX
Mailbox implementation for the Broadcom PDC ring manager,
which provides access to various offload engines on Broadcom
SoCs. Say Y here if you want to use the Broadcom PDC.
+
+config BCM_FLEXRM_MBOX
+ tristate "Broadcom FlexRM Mailbox"
+ depends on ARM64 || COMPILE_TEST
+ depends on HAS_DMA
+ select GENERIC_MSI_IRQ_DOMAIN
+ default ARCH_BCM_IPROC
+ help
+ Mailbox implementation of the Broadcom FlexRM ring manager,
+ which provides access to various offload engines on Broadcom
+ SoCs. Say Y here if you want to use the Broadcom FlexRM.
endif
diff --git a/drivers/mailbox/Makefile b/drivers/mailbox/Makefile
index 7dde4f6..45083c0 100644
--- a/drivers/mailbox/Makefile
+++ b/drivers/mailbox/Makefile
@@ -30,4 +30,6 @@ obj-$(CONFIG_HI6220_MBOX) += hi6220-mailbox.o
obj-$(CONFIG_BCM_PDC_MBOX) += bcm-pdc-mailbox.o
+obj-$(CONFIG_BCM_FLEXRM_MBOX) += mailbox-flexrm/
+
obj-$(CONFIG_TEGRA_HSP_MBOX) += tegra-hsp.o
diff --git a/drivers/mailbox/mailbox-flexrm/Makefile b/drivers/mailbox/mailbox-flexrm/Makefile
new file mode 100644
index 0000000..f5bf069
--- /dev/null
+++ b/drivers/mailbox/mailbox-flexrm/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for Broadcom FlexRM Mailbox Driver.
+#
+
+flexrm-mbox-objs := flexrm-main.o flexrm-desc.o
+obj-$(CONFIG_BCM_FLEXRM_MBOX) += flexrm-mbox.o
diff --git a/drivers/mailbox/mailbox-flexrm/flexrm-desc.c b/drivers/mailbox/mailbox-flexrm/flexrm-desc.c
new file mode 100644
index 0000000..b0449eb
--- /dev/null
+++ b/drivers/mailbox/mailbox-flexrm/flexrm-desc.c
@@ -0,0 +1,764 @@
+/* Broadcom FlexRM Mailbox Driver
+ *
+ * Copyright (C) 2016 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * FlexRM descriptor library
+ */
+
+#include <asm/barrier.h>
+#include <asm/byteorder.h>
+#include <linux/dma-mapping.h>
+#include <linux/printk.h>
+
+#include "flexrm-desc.h"
+
+/* Completion descriptor format */
+#define CMPL_OPAQUE_SHIFT 0
+#define CMPL_OPAQUE_MASK 0xffff
+#define CMPL_ENGINE_STATUS_SHIFT 16
+#define CMPL_ENGINE_STATUS_MASK 0xffff
+#define CMPL_DME_STATUS_SHIFT 32
+#define CMPL_DME_STATUS_MASK 0xffff
+#define CMPL_RM_STATUS_SHIFT 48
+#define CMPL_RM_STATUS_MASK 0xffff
+
+/* Completion DME status code */
+#define DME_STATUS_MEM_COR_ERR BIT(0)
+#define DME_STATUS_MEM_UCOR_ERR BIT(1)
+#define DME_STATUS_FIFO_UNDERFLOW BIT(2)
+#define DME_STATUS_FIFO_OVERFLOW BIT(3)
+#define DME_STATUS_RRESP_ERR BIT(4)
+#define DME_STATUS_BRESP_ERR BIT(5)
+#define DME_STATUS_ERROR_MASK (DME_STATUS_MEM_COR_ERR | \
+ DME_STATUS_MEM_UCOR_ERR | \
+ DME_STATUS_FIFO_UNDERFLOW | \
+ DME_STATUS_FIFO_OVERFLOW | \
+ DME_STATUS_RRESP_ERR | \
+ DME_STATUS_BRESP_ERR)
+
+/* Completion RM status code */
+#define RM_STATUS_CODE_SHIFT 0
+#define RM_STATUS_CODE_MASK 0x3ff
+#define RM_STATUS_CODE_GOOD 0x0
+#define RM_STATUS_CODE_AE_TIMEOUT 0x3ff
+
+/* General descriptor format */
+#define DESC_TYPE_SHIFT 60
+#define DESC_TYPE_MASK 0xf
+#define DESC_PAYLOAD_SHIFT 0
+#define DESC_PAYLOAD_MASK 0x0fffffffffffffff
+
+/* Null descriptor format */
+#define NULL_TYPE 0
+#define NULL_TOGGLE_SHIFT 58
+#define NULL_TOGGLE_MASK 0x1
+
+/* Header descriptor format */
+#define HEADER_TYPE 1
+#define HEADER_TOGGLE_SHIFT 58
+#define HEADER_TOGGLE_MASK 0x1
+#define HEADER_ENDPKT_SHIFT 57
+#define HEADER_ENDPKT_MASK 0x1
+#define HEADER_STARTPKT_SHIFT 56
+#define HEADER_STARTPKT_MASK 0x1
+#define HEADER_BDCOUNT_SHIFT 36
+#define HEADER_BDCOUNT_MASK 0x1f
+#define HEADER_BDCOUNT_MAX HEADER_BDCOUNT_MASK
+#define HEADER_FLAGS_SHIFT 16
+#define HEADER_FLAGS_MASK 0xffff
+#define HEADER_OPAQUE_SHIFT 0
+#define HEADER_OPAQUE_MASK 0xffff
+
+/* Source (SRC) descriptor format */
+#define SRC_TYPE 2
+#define SRC_LENGTH_SHIFT 44
+#define SRC_LENGTH_MASK 0xffff
+#define SRC_ADDR_SHIFT 0
+#define SRC_ADDR_MASK 0x00000fffffffffff
+
+/* Destination (DST) descriptor format */
+#define DST_TYPE 3
+#define DST_LENGTH_SHIFT 44
+#define DST_LENGTH_MASK 0xffff
+#define DST_ADDR_SHIFT 0
+#define DST_ADDR_MASK 0x00000fffffffffff
+
+/* Immediate (IMM) descriptor format */
+#define IMM_TYPE 4
+#define IMM_DATA_SHIFT 0
+#define IMM_DATA_MASK 0x0fffffffffffffff
+
+/* Next pointer (NPTR) descriptor format */
+#define NPTR_TYPE 5
+#define NPTR_TOGGLE_SHIFT 58
+#define NPTR_TOGGLE_MASK 0x1
+#define NPTR_ADDR_SHIFT 0
+#define NPTR_ADDR_MASK 0x00000fffffffffff
+
+/* Mega source (MSRC) descriptor format */
+#define MSRC_TYPE 6
+#define MSRC_LENGTH_SHIFT 44
+#define MSRC_LENGTH_MASK 0xffff
+#define MSRC_ADDR_SHIFT 0
+#define MSRC_ADDR_MASK 0x00000fffffffffff
+
+/* Mega destination (MDST) descriptor format */
+#define MDST_TYPE 7
+#define MDST_LENGTH_SHIFT 44
+#define MDST_LENGTH_MASK 0xffff
+#define MDST_ADDR_SHIFT 0
+#define MDST_ADDR_MASK 0x00000fffffffffff
+
+/* Source with tlast (SRCT) descriptor format */
+#define SRCT_TYPE 8
+#define SRCT_LENGTH_SHIFT 44
+#define SRCT_LENGTH_MASK 0xffff
+#define SRCT_ADDR_SHIFT 0
+#define SRCT_ADDR_MASK 0x00000fffffffffff
+
+/* Destination with tlast (DSTT) descriptor format */
+#define DSTT_TYPE 9
+#define DSTT_LENGTH_SHIFT 44
+#define DSTT_LENGTH_MASK 0xffff
+#define DSTT_ADDR_SHIFT 0
+#define DSTT_ADDR_MASK 0x00000fffffffffff
+
+/* Immediate with tlast (IMMT) descriptor format */
+#define IMMT_TYPE 10
+#define IMMT_DATA_SHIFT 0
+#define IMMT_DATA_MASK 0x0fffffffffffffff
+
+/* Descriptor helper macros */
+#define DESC_DEC(_d, _s, _m) (((_d) >> (_s)) & (_m))
+#define DESC_ENC(_d, _v, _s, _m) \
+ do { \
+ (_d) &= ~((u64)(_m) << (_s)); \
+ (_d) |= (((u64)(_v) & (_m)) << (_s)); \
+ } while (0)
+
+u64 flexrm_read_desc(void *desc_ptr)
+{
+ return le64_to_cpu(*((u64 *)desc_ptr));
+}
+
+void flexrm_write_desc(void *desc_ptr, u64 desc)
+{
+ *((u64 *)desc_ptr) = cpu_to_le64(desc);
+}
+
+u32 flexrm_cmpl_desc_to_reqid(u64 cmpl_desc)
+{
+ return (u32)(cmpl_desc & CMPL_OPAQUE_MASK);
+}
+
+int flexrm_cmpl_desc_to_error(u64 cmpl_desc)
+{
+ u32 status;
+
+ status = DESC_DEC(cmpl_desc, CMPL_DME_STATUS_SHIFT,
+ CMPL_DME_STATUS_MASK);
+ if (status & DME_STATUS_ERROR_MASK)
+ return -EIO;
+
+ status = DESC_DEC(cmpl_desc, CMPL_RM_STATUS_SHIFT,
+ CMPL_RM_STATUS_MASK);
+ status &= RM_STATUS_CODE_MASK;
+ if (status == RM_STATUS_CODE_AE_TIMEOUT)
+ return -ETIMEDOUT;
+
+ return 0;
+}
+
+bool flexrm_is_next_table_desc(void *desc_ptr)
+{
+ u64 desc = flexrm_read_desc(desc_ptr);
+ u32 type = DESC_DEC(desc, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+
+ return (type == NPTR_TYPE) ? true : false;
+}
+
+u64 flexrm_next_table_desc(u32 toggle, dma_addr_t next_addr)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, NPTR_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, toggle, NPTR_TOGGLE_SHIFT, NPTR_TOGGLE_MASK);
+ DESC_ENC(desc, next_addr, NPTR_ADDR_SHIFT, NPTR_ADDR_MASK);
+
+ return desc;
+}
+
+u64 flexrm_null_desc(u32 toggle)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, NULL_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, toggle, NULL_TOGGLE_SHIFT, NULL_TOGGLE_MASK);
+
+ return desc;
+}
+
+u32 flexrm_estimate_header_desc_count(u32 nhcnt)
+{
+ u32 hcnt = nhcnt / HEADER_BDCOUNT_MAX;
+
+ if (!(nhcnt % HEADER_BDCOUNT_MAX))
+ hcnt += 1;
+
+ return hcnt;
+}
+
+static void flexrm_flip_header_toogle(void *desc_ptr)
+{
+ u64 desc = flexrm_read_desc(desc_ptr);
+
+ if (desc & ((u64)0x1 << HEADER_TOGGLE_SHIFT))
+ desc &= ~((u64)0x1 << HEADER_TOGGLE_SHIFT);
+ else
+ desc |= ((u64)0x1 << HEADER_TOGGLE_SHIFT);
+
+ flexrm_write_desc(desc_ptr, desc);
+}
+
+static u64 flexrm_header_desc(u32 toggle, u32 startpkt, u32 endpkt,
+ u32 bdcount, u32 flags, u32 opaque)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, HEADER_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, toggle, HEADER_TOGGLE_SHIFT, HEADER_TOGGLE_MASK);
+ DESC_ENC(desc, startpkt, HEADER_STARTPKT_SHIFT, HEADER_STARTPKT_MASK);
+ DESC_ENC(desc, endpkt, HEADER_ENDPKT_SHIFT, HEADER_ENDPKT_MASK);
+ DESC_ENC(desc, bdcount, HEADER_BDCOUNT_SHIFT, HEADER_BDCOUNT_MASK);
+ DESC_ENC(desc, flags, HEADER_FLAGS_SHIFT, HEADER_FLAGS_MASK);
+ DESC_ENC(desc, opaque, HEADER_OPAQUE_SHIFT, HEADER_OPAQUE_MASK);
+
+ return desc;
+}
+
+static void flexrm_enqueue_desc(u32 nhpos, u32 nhcnt, u32 reqid,
+ u64 desc, void **desc_ptr, u32 *toggle,
+ void *start_desc, void *end_desc)
+{
+ u64 d;
+ u32 nhavail, _toggle, _startpkt, _endpkt, _bdcount;
+
+ /* Sanity check */
+ if (nhcnt <= nhpos)
+ return;
+
+ /*
+ * Each request or packet start with a HEADER descriptor followed
+ * by one or more non-HEADER descriptors (SRC, SRCT, MSRC, DST,
+ * DSTT, MDST, IMM, and IMMT). The number of non-HEADER descriptors
+ * following a HEADER descriptor is represented by BDCOUNT field
+ * of HEADER descriptor. The max value of BDCOUNT field is 31 which
+ * means we can only have 31 non-HEADER descriptors following one
+ * HEADER descriptor.
+ *
+ * In general use, number of non-HEADER descriptors can easily go
+ * beyond 31. To tackle this situation, we have packet (or request)
+ * extenstion bits (STARTPKT and ENDPKT) in the HEADER descriptor.
+ *
+ * To use packet extension, the first HEADER descriptor of request
+ * (or packet) will have STARTPKT=1 and ENDPKT=0. The intermediate
+ * HEADER descriptors will have STARTPKT=0 and ENDPKT=0. The last
+ * HEADER descriptor will have STARTPKT=0 and ENDPKT=1. Also, the
+ * TOGGLE bit of the first HEADER will be set to invalid state to
+ * ensure that FlexRM does not start fetching descriptors till all
+ * descriptors are enqueued. The user of this function will flip
+ * the TOGGLE bit of first HEADER after all descriptors are
+ * enqueued.
+ */
+
+ if ((nhpos % HEADER_BDCOUNT_MAX == 0) && (nhcnt - nhpos)) {
+ /* Prepare the header descriptor */
+ nhavail = (nhcnt - nhpos);
+ _toggle = (nhpos == 0) ? !(*toggle) : (*toggle);
+ _startpkt = (nhpos == 0) ? 0x1 : 0x0;
+ _endpkt = (nhavail <= HEADER_BDCOUNT_MAX) ? 0x1 : 0x0;
+ _bdcount = (nhavail <= HEADER_BDCOUNT_MAX) ?
+ nhavail : HEADER_BDCOUNT_MAX;
+ if (nhavail <= HEADER_BDCOUNT_MAX)
+ _bdcount = nhavail;
+ else
+ _bdcount = HEADER_BDCOUNT_MAX;
+ d = flexrm_header_desc(_toggle, _startpkt, _endpkt,
+ _bdcount, 0x0, reqid);
+
+ /* Write header descriptor */
+ flexrm_write_desc(*desc_ptr, d);
+
+ /* Point to next descriptor */
+ *desc_ptr += sizeof(desc);
+ if (*desc_ptr == end_desc)
+ *desc_ptr = start_desc;
+
+ /* Skip next pointer descriptors */
+ while (flexrm_is_next_table_desc(*desc_ptr)) {
+ *toggle = (*toggle) ? 0 : 1;
+ *desc_ptr += sizeof(desc);
+ if (*desc_ptr == end_desc)
+ *desc_ptr = start_desc;
+ }
+ }
+
+ /* Write desired descriptor */
+ flexrm_write_desc(*desc_ptr, desc);
+
+ /* Point to next descriptor */
+ *desc_ptr += sizeof(desc);
+ if (*desc_ptr == end_desc)
+ *desc_ptr = start_desc;
+
+ /* Skip next pointer descriptors */
+ while (flexrm_is_next_table_desc(*desc_ptr)) {
+ *toggle = (*toggle) ? 0 : 1;
+ *desc_ptr += sizeof(desc);
+ if (*desc_ptr == end_desc)
+ *desc_ptr = start_desc;
+ }
+}
+
+static u64 flexrm_src_desc(dma_addr_t addr, unsigned int length)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, SRC_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, length, SRC_LENGTH_SHIFT, SRC_LENGTH_MASK);
+ DESC_ENC(desc, addr, SRC_ADDR_SHIFT, SRC_ADDR_MASK);
+
+ return desc;
+}
+
+static u64 flexrm_msrc_desc(dma_addr_t addr, unsigned int length_div_16)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, MSRC_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, length_div_16, MSRC_LENGTH_SHIFT, MSRC_LENGTH_MASK);
+ DESC_ENC(desc, addr, MSRC_ADDR_SHIFT, MSRC_ADDR_MASK);
+
+ return desc;
+}
+
+static u64 flexrm_dst_desc(dma_addr_t addr, unsigned int length)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, DST_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, length, DST_LENGTH_SHIFT, DST_LENGTH_MASK);
+ DESC_ENC(desc, addr, DST_ADDR_SHIFT, DST_ADDR_MASK);
+
+ return desc;
+}
+
+static u64 flexrm_mdst_desc(dma_addr_t addr, unsigned int length_div_16)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, MDST_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, length_div_16, MDST_LENGTH_SHIFT, MDST_LENGTH_MASK);
+ DESC_ENC(desc, addr, MDST_ADDR_SHIFT, MDST_ADDR_MASK);
+
+ return desc;
+}
+
+static u64 flexrm_imm_desc(u64 data)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, IMM_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, data, IMM_DATA_SHIFT, IMM_DATA_MASK);
+
+ return desc;
+}
+
+static u64 flexrm_srct_desc(dma_addr_t addr, unsigned int length)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, SRCT_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, length, SRCT_LENGTH_SHIFT, SRCT_LENGTH_MASK);
+ DESC_ENC(desc, addr, SRCT_ADDR_SHIFT, SRCT_ADDR_MASK);
+
+ return desc;
+}
+
+static u64 flexrm_dstt_desc(dma_addr_t addr, unsigned int length)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, DSTT_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, length, DSTT_LENGTH_SHIFT, DSTT_LENGTH_MASK);
+ DESC_ENC(desc, addr, DSTT_ADDR_SHIFT, DSTT_ADDR_MASK);
+
+ return desc;
+}
+
+static u64 flexrm_immt_desc(u64 data)
+{
+ u64 desc = 0;
+
+ DESC_ENC(desc, IMMT_TYPE, DESC_TYPE_SHIFT, DESC_TYPE_MASK);
+ DESC_ENC(desc, data, IMMT_DATA_SHIFT, IMMT_DATA_MASK);
+
+ return desc;
+}
+
+static bool flexrm_spu_sanity_check(struct brcm_message *msg)
+{
+ struct scatterlist *sg;
+
+ if (!msg->spu.src || !msg->spu.dst)
+ return false;
+ for (sg = msg->spu.src; sg; sg = sg_next(sg)) {
+ if (sg->length & 0xf) {
+ if (sg->length > SRC_LENGTH_MASK)
+ return false;
+ } else {
+ if (sg->length > (MSRC_LENGTH_MASK * 16))
+ return false;
+ }
+ }
+ for (sg = msg->spu.dst; sg; sg = sg_next(sg)) {
+ if (sg->length & 0xf) {
+ if (sg->length > DST_LENGTH_MASK)
+ return false;
+ } else {
+ if (sg->length > (MDST_LENGTH_MASK * 16))
+ return false;
+ }
+ }
+
+ return true;
+}
+
+static u32 flexrm_spu_estimate_nonheader_desc_count(struct brcm_message *msg)
+{
+ u32 cnt = 0;
+ unsigned int dst_target = 0;
+ struct scatterlist *src_sg = msg->spu.src, *dst_sg = msg->spu.dst;
+
+ while (src_sg || dst_sg) {
+ if (src_sg) {
+ cnt++;
+ dst_target = src_sg->length;
+ src_sg = sg_next(src_sg);
+ } else
+ dst_target = UINT_MAX;
+
+ while (dst_target && dst_sg) {
+ cnt++;
+ if (dst_sg->length < dst_target)
+ dst_target -= dst_sg->length;
+ else
+ dst_target = 0;
+ dst_sg = sg_next(dst_sg);
+ }
+ }
+
+ return cnt;
+}
+
+static int flexrm_spu_dma_map(struct device *dev, struct brcm_message *msg)
+{
+ int rc;
+
+ rc = dma_map_sg(dev, msg->spu.src, sg_nents(msg->spu.src),
+ DMA_TO_DEVICE);
+ if (rc < 0)
+ return rc;
+
+ rc = dma_map_sg(dev, msg->spu.dst, sg_nents(msg->spu.dst),
+ DMA_FROM_DEVICE);
+ if (rc < 0) {
+ dma_unmap_sg(dev, msg->spu.src, sg_nents(msg->spu.src),
+ DMA_TO_DEVICE);
+ return rc;
+ }
+
+ return 0;
+}
+
+static void flexrm_spu_dma_unmap(struct device *dev, struct brcm_message *msg)
+{
+ dma_unmap_sg(dev, msg->spu.dst, sg_nents(msg->spu.dst),
+ DMA_FROM_DEVICE);
+ dma_unmap_sg(dev, msg->spu.src, sg_nents(msg->spu.src),
+ DMA_TO_DEVICE);
+}
+
+static void *flexrm_spu_write_descs(struct brcm_message *msg, u32 nhcnt,
+ u32 reqid, void *desc_ptr, u32 toggle,
+ void *start_desc, void *end_desc)
+{
+ u64 d;
+ u32 nhpos = 0;
+ void *orig_desc_ptr = desc_ptr;
+ unsigned int dst_target = 0;
+ struct scatterlist *src_sg = msg->spu.src, *dst_sg = msg->spu.dst;
+
+ while (src_sg || dst_sg) {
+ if (src_sg) {
+ if (sg_dma_len(src_sg) & 0xf)
+ d = flexrm_src_desc(sg_dma_address(src_sg),
+ sg_dma_len(src_sg));
+ else
+ d = flexrm_msrc_desc(sg_dma_address(src_sg),
+ sg_dma_len(src_sg)/16);
+ flexrm_enqueue_desc(nhpos, nhcnt, reqid,
+ d, &desc_ptr, &toggle,
+ start_desc, end_desc);
+ nhpos++;
+ dst_target = sg_dma_len(src_sg);
+ src_sg = sg_next(src_sg);
+ } else
+ dst_target = UINT_MAX;
+
+ while (dst_target && dst_sg) {
+ if (sg_dma_len(dst_sg) & 0xf)
+ d = flexrm_dst_desc(sg_dma_address(dst_sg),
+ sg_dma_len(dst_sg));
+ else
+ d = flexrm_mdst_desc(sg_dma_address(dst_sg),
+ sg_dma_len(dst_sg)/16);
+ flexrm_enqueue_desc(nhpos, nhcnt, reqid,
+ d, &desc_ptr, &toggle,
+ start_desc, end_desc);
+ nhpos++;
+ if (sg_dma_len(dst_sg) < dst_target)
+ dst_target -= sg_dma_len(dst_sg);
+ else
+ dst_target = 0;
+ dst_sg = sg_next(dst_sg);
+ }
+ }
+
+ /* Null descriptor with invalid toggle bit */
+ flexrm_write_desc(desc_ptr, flexrm_null_desc(!toggle));
+
+ /* Ensure that descriptors have been written to memory */
+ wmb();
+
+ /* Flip toggle bit in header */
+ flexrm_flip_header_toogle(orig_desc_ptr);
+
+ return desc_ptr;
+}
+
+static bool flexrm_sba_sanity_check(struct brcm_message *msg)
+{
+ u32 i;
+
+ if (!msg->sba.cmds || !msg->sba.cmds_count)
+ return false;
+
+ for (i = 0; i < msg->sba.cmds_count; i++) {
+ if (((msg->sba.cmds[i].flags & BRCM_SBA_CMD_TYPE_B) ||
+ (msg->sba.cmds[i].flags & BRCM_SBA_CMD_TYPE_C)) &&
+ (msg->sba.cmds[i].flags & BRCM_SBA_CMD_HAS_OUTPUT))
+ return false;
+ if ((msg->sba.cmds[i].flags & BRCM_SBA_CMD_TYPE_B) &&
+ (msg->sba.cmds[i].data_len > SRCT_LENGTH_MASK))
+ return false;
+ if ((msg->sba.cmds[i].flags & BRCM_SBA_CMD_TYPE_C) &&
+ (msg->sba.cmds[i].data_len > SRCT_LENGTH_MASK))
+ return false;
+ if ((msg->sba.cmds[i].flags & BRCM_SBA_CMD_HAS_RESP) &&
+ (msg->sba.cmds[i].resp_len > DSTT_LENGTH_MASK))
+ return false;
+ if ((msg->sba.cmds[i].flags & BRCM_SBA_CMD_HAS_OUTPUT) &&
+ (msg->sba.cmds[i].data_len > DSTT_LENGTH_MASK))
+ return false;
+ }
+
+ return true;
+}
+
+static u32 flexrm_sba_estimate_nonheader_desc_count(struct brcm_message *msg)
+{
+ u32 i, cnt;
+
+ cnt = 0;
+ for (i = 0; i < msg->sba.cmds_count; i++) {
+ cnt++;
+
+ if ((msg->sba.cmds[i].flags & BRCM_SBA_CMD_TYPE_B) ||
+ (msg->sba.cmds[i].flags & BRCM_SBA_CMD_TYPE_C))
+ cnt++;
+
+ if (msg->sba.cmds[i].flags & BRCM_SBA_CMD_HAS_RESP)
+ cnt++;
+
+ if (msg->sba.cmds[i].flags & BRCM_SBA_CMD_HAS_OUTPUT)
+ cnt++;
+ }
+
+ return cnt;
+}
+
+static void *flexrm_sba_write_descs(struct brcm_message *msg, u32 nhcnt,
+ u32 reqid, void *desc_ptr, u32 toggle,
+ void *start_desc, void *end_desc)
+{
+ u64 d;
+ u32 i, nhpos = 0;
+ struct brcm_sba_command *c;
+ void *orig_desc_ptr = desc_ptr;
+
+ /* Convert SBA commands into descriptors */
+ for (i = 0; i < msg->sba.cmds_count; i++) {
+ c = &msg->sba.cmds[i];
+
+ if ((c->flags & BRCM_SBA_CMD_HAS_RESP) &&
+ (c->flags & BRCM_SBA_CMD_HAS_OUTPUT)) {
+ /* Destination response descriptor */
+ d = flexrm_dst_desc(c->resp, c->resp_len);
+ flexrm_enqueue_desc(nhpos, nhcnt, reqid,
+ d, &desc_ptr, &toggle,
+ start_desc, end_desc);
+ nhpos++;
+ } else if (c->flags & BRCM_SBA_CMD_HAS_RESP) {
+ /* Destination response with tlast descriptor */
+ d = flexrm_dstt_desc(c->resp, c->resp_len);
+ flexrm_enqueue_desc(nhpos, nhcnt, reqid,
+ d, &desc_ptr, &toggle,
+ start_desc, end_desc);
+ nhpos++;
+ }
+
+ if (c->flags & BRCM_SBA_CMD_HAS_OUTPUT) {
+ /* Destination with tlast descriptor */
+ d = flexrm_dstt_desc(c->data, c->data_len);
+ flexrm_enqueue_desc(nhpos, nhcnt, reqid,
+ d, &desc_ptr, &toggle,
+ start_desc, end_desc);
+ nhpos++;
+ }
+
+ if (c->flags & BRCM_SBA_CMD_TYPE_B) {
+ /* Command as immediate descriptor */
+ d = flexrm_imm_desc(c->cmd);
+ flexrm_enqueue_desc(nhpos, nhcnt, reqid,
+ d, &desc_ptr, &toggle,
+ start_desc, end_desc);
+ nhpos++;
+ } else {
+ /* Command as immediate descriptor with tlast */
+ d = flexrm_immt_desc(c->cmd);
+ flexrm_enqueue_desc(nhpos, nhcnt, reqid,
+ d, &desc_ptr, &toggle,
+ start_desc, end_desc);
+ nhpos++;
+ }
+
+ if ((c->flags & BRCM_SBA_CMD_TYPE_B) ||
+ (c->flags & BRCM_SBA_CMD_TYPE_C)) {
+ /* Source with tlast descriptor */
+ d = flexrm_srct_desc(c->data, c->data_len);
+ flexrm_enqueue_desc(nhpos, nhcnt, reqid,
+ d, &desc_ptr, &toggle,
+ start_desc, end_desc);
+ nhpos++;
+ }
+ }
+
+ /* Null descriptor with invalid toggle bit */
+ flexrm_write_desc(desc_ptr, flexrm_null_desc(!toggle));
+
+ /* Ensure that descriptors have been written to memory */
+ wmb();
+
+ /* Flip toggle bit in header */
+ flexrm_flip_header_toogle(orig_desc_ptr);
+
+ return desc_ptr;
+}
+
+bool flexrm_sanity_check(struct brcm_message *msg)
+{
+ if (!msg)
+ return false;
+
+ switch (msg->type) {
+ case BRCM_MESSAGE_SPU:
+ return flexrm_spu_sanity_check(msg);
+ case BRCM_MESSAGE_SBA:
+ return flexrm_sba_sanity_check(msg);
+ default:
+ return false;
+ };
+}
+
+u32 flexrm_estimate_nonheader_desc_count(struct brcm_message *msg)
+{
+ if (!msg)
+ return 0;
+
+ switch (msg->type) {
+ case BRCM_MESSAGE_SPU:
+ return flexrm_spu_estimate_nonheader_desc_count(msg);
+ case BRCM_MESSAGE_SBA:
+ return flexrm_sba_estimate_nonheader_desc_count(msg);
+ default:
+ return 0;
+ };
+}
+
+int flexrm_dma_map(struct device *dev, struct brcm_message *msg)
+{
+ if (!dev || !msg)
+ return -EINVAL;
+
+ switch (msg->type) {
+ case BRCM_MESSAGE_SPU:
+ return flexrm_spu_dma_map(dev, msg);
+ default:
+ break;
+ }
+
+ return 0;
+}
+
+void flexrm_dma_unmap(struct device *dev, struct brcm_message *msg)
+{
+ if (!dev || !msg)
+ return;
+
+ switch (msg->type) {
+ case BRCM_MESSAGE_SPU:
+ flexrm_spu_dma_unmap(dev, msg);
+ break;
+ default:
+ break;
+ }
+}
+
+void *flexrm_write_descs(struct brcm_message *msg, u32 nhcnt,
+ u32 reqid, void *desc_ptr, u32 toggle,
+ void *start_desc, void *end_desc)
+{
+ if (!msg || !desc_ptr || !start_desc || !end_desc)
+ return ERR_PTR(-ENOTSUPP);
+
+ if ((desc_ptr < start_desc) || (end_desc <= desc_ptr))
+ return ERR_PTR(-ERANGE);
+
+ switch (msg->type) {
+ case BRCM_MESSAGE_SPU:
+ return flexrm_spu_write_descs(msg, nhcnt, reqid,
+ desc_ptr, toggle,
+ start_desc, end_desc);
+ case BRCM_MESSAGE_SBA:
+ return flexrm_sba_write_descs(msg, nhcnt, reqid,
+ desc_ptr, toggle,
+ start_desc, end_desc);
+ default:
+ return ERR_PTR(-ENOTSUPP);
+ };
+}
diff --git a/drivers/mailbox/mailbox-flexrm/flexrm-desc.h b/drivers/mailbox/mailbox-flexrm/flexrm-desc.h
new file mode 100644
index 0000000..a95cf61
--- /dev/null
+++ b/drivers/mailbox/mailbox-flexrm/flexrm-desc.h
@@ -0,0 +1,47 @@
+/* Broadcom FlexRM Mailbox Driver
+ *
+ * Copyright (C) 2016 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * FlexRM descriptor library
+ */
+
+#ifndef __FLEXRM_DESC_H__
+#define __FLEXRM_DESC_H__
+
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/mailbox/brcm-message.h>
+
+extern u64 flexrm_read_desc(void *desc_ptr);
+
+extern void flexrm_write_desc(void *desc_ptr, u64 desc);
+
+extern u32 flexrm_cmpl_desc_to_reqid(u64 cmpl_desc);
+
+extern int flexrm_cmpl_desc_to_error(u64 cmpl_desc);
+
+extern bool flexrm_is_next_table_desc(void *desc_ptr);
+
+extern u64 flexrm_next_table_desc(u32 toggle, dma_addr_t next_addr);
+
+extern u64 flexrm_null_desc(u32 toogle);
+
+extern u32 flexrm_estimate_header_desc_count(u32 nhcnt);
+
+extern bool flexrm_sanity_check(struct brcm_message *msg);
+
+extern u32 flexrm_estimate_nonheader_desc_count(struct brcm_message *msg);
+
+extern int flexrm_dma_map(struct device *dev, struct brcm_message *msg);
+
+extern void flexrm_dma_unmap(struct device *dev, struct brcm_message *msg);
+
+extern void *flexrm_write_descs(struct brcm_message *msg, u32 nhcnt,
+ u32 reqid, void *desc_ptr, u32 toggle,
+ void *start_desc, void *end_desc);
+
+#endif /* __FLEXRM_DESC_H__ */
diff --git a/drivers/mailbox/mailbox-flexrm/flexrm-main.c b/drivers/mailbox/mailbox-flexrm/flexrm-main.c
new file mode 100644
index 0000000..c8890f1
--- /dev/null
+++ b/drivers/mailbox/mailbox-flexrm/flexrm-main.c
@@ -0,0 +1,829 @@
+/* Broadcom FlexRM Mailbox Driver
+ *
+ * Copyright (C) 2016 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Each Broadcom FlexSparx4 offload engine is implemented as an
+ * extension to Broadcom FlexRM ring manager. The FlexRM ring
+ * manager provides a set of rings which can be used to submit
+ * work to a FlexSparx4 offload engine.
+ *
+ * This driver creates a mailbox controller using a set of FlexRM
+ * rings where each mailbox channel represents a separate FlexRM ring.
+ */
+
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmapool.h>
+#include <linux/err.h>
+#include <linux/idr.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/mailbox_controller.h>
+#include <linux/mailbox_client.h>
+#include <linux/mailbox/brcm-message.h>
+#include <linux/module.h>
+#include <linux/msi.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/platform_device.h>
+#include <linux/spinlock.h>
+
+#include "flexrm-desc.h"
+
+/* FlexRM configuration */
+#define RING_REGS_SIZE 0x10000
+#define RING_DESC_SIZE 8
+#define RING_DESC_INDEX(offset) \
+ ((offset) / RING_DESC_SIZE)
+#define RING_DESC_OFFSET(index) \
+ ((index) * RING_DESC_SIZE)
+#define RING_MAX_REQ_COUNT 1024
+#define RING_BD_ALIGN_ORDER 12
+#define RING_BD_ALIGN_CHECK(addr) \
+ (!((addr) & ((0x1 << RING_BD_ALIGN_ORDER) - 1)))
+#define RING_BD_TOGGLE_INVALID(offset) \
+ (((offset) >> RING_BD_ALIGN_ORDER) & 0x1)
+#define RING_BD_TOGGLE_VALID(offset) \
+ (!RING_BD_TOGGLE_INVALID(offset))
+#define RING_BD_DESC_PER_REQ 32
+#define RING_BD_DESC_COUNT \
+ (RING_MAX_REQ_COUNT * RING_BD_DESC_PER_REQ)
+#define RING_BD_SIZE \
+ (RING_BD_DESC_COUNT * RING_DESC_SIZE)
+#define RING_CMPL_ALIGN_ORDER 13
+#define RING_CMPL_DESC_COUNT RING_MAX_REQ_COUNT
+#define RING_CMPL_SIZE \
+ (RING_CMPL_DESC_COUNT * RING_DESC_SIZE)
+#define RING_VER_MAGIC 0x76303031
+
+/* Per-Ring register offsets */
+#define RING_VER 0x000
+#define RING_BD_START_ADDR 0x004
+#define RING_BD_READ_PTR 0x008
+#define RING_BD_WRITE_PTR 0x00c
+#define RING_BD_READ_PTR_DDR_LS 0x010
+#define RING_BD_READ_PTR_DDR_MS 0x014
+#define RING_CMPL_START_ADDR 0x018
+#define RING_CMPL_WRITE_PTR 0x01c
+#define RING_NUM_REQ_RECV_LS 0x020
+#define RING_NUM_REQ_RECV_MS 0x024
+#define RING_NUM_REQ_TRANS_LS 0x028
+#define RING_NUM_REQ_TRANS_MS 0x02c
+#define RING_NUM_REQ_OUTSTAND 0x030
+#define RING_CONTROL 0x034
+#define RING_FLUSH_DONE 0x038
+#define RING_MSI_ADDR_LS 0x03c
+#define RING_MSI_ADDR_MS 0x040
+#define RING_MSI_CONTROL 0x048
+#define RING_BD_READ_PTR_DDR_CONTROL 0x04c
+#define RING_MSI_DATA_VALUE 0x064
+
+/* Register RING_BD_START_ADDR fields */
+#define BD_LAST_UPDATE_HW_SHIFT 28
+#define BD_LAST_UPDATE_HW_MASK 0x1
+#define BD_START_ADDR_VALUE(pa) \
+ ((u32)((((dma_addr_t)(pa)) >> RING_BD_ALIGN_ORDER) & 0x0fffffff))
+#define BD_START_ADDR_DECODE(val) \
+ ((dma_addr_t)((val) & 0x0fffffff) << RING_BD_ALIGN_ORDER)
+
+/* Register RING_CMPL_START_ADDR fields */
+#define CMPL_START_ADDR_VALUE(pa) \
+ ((u32)((((u64)(pa)) >> RING_CMPL_ALIGN_ORDER) & 0x03ffffff))
+
+/* Register RING_CONTROL fields */
+#define CONTROL_MASK_DISABLE_CONTROL 12
+#define CONTROL_FLUSH_SHIFT 5
+#define CONTROL_ACTIVE_SHIFT 4
+#define CONTROL_RATE_ADAPT_MASK 0xf
+#define CONTROL_RATE_DYNAMIC 0x0
+#define CONTROL_RATE_FAST 0x8
+#define CONTROL_RATE_MEDIUM 0x9
+#define CONTROL_RATE_SLOW 0xa
+#define CONTROL_RATE_IDLE 0xb
+
+/* Register RING_FLUSH_DONE fields */
+#define FLUSH_DONE_MASK 0x1
+
+/* Register RING_MSI_CONTROL fields */
+#define MSI_TIMER_VAL_SHIFT 16
+#define MSI_TIMER_VAL_MASK 0xffff
+#define MSI_ENABLE_SHIFT 15
+#define MSI_ENABLE_MASK 0x1
+#define MSI_COUNT_SHIFT 0
+#define MSI_COUNT_MASK 0x3ff
+
+/* Register RING_BD_READ_PTR_DDR_CONTROL fields */
+#define BD_READ_PTR_DDR_TIMER_VAL_SHIFT 16
+#define BD_READ_PTR_DDR_TIMER_VAL_MASK 0xffff
+#define BD_READ_PTR_DDR_ENABLE_SHIFT 15
+#define BD_READ_PTR_DDR_ENABLE_MASK 0x1
+
+struct flexrm_ring {
+ /* Unprotected members */
+ int num;
+ struct flexrm_mbox *mbox;
+ void __iomem *regs;
+ bool irq_requested;
+ unsigned int irq;
+ unsigned int msi_timer_val;
+ unsigned int msi_count_threshold;
+ struct ida requests_ida;
+ struct brcm_message *requests[RING_MAX_REQ_COUNT];
+ void *bd_base;
+ dma_addr_t bd_dma_base;
+ u32 bd_write_offset;
+ void *cmpl_base;
+ dma_addr_t cmpl_dma_base;
+ /* Protected members */
+ spinlock_t lock;
+ struct brcm_message *last_pending_msg;
+ u32 cmpl_read_offset;
+};
+
+struct flexrm_mbox {
+ struct device *dev;
+ void __iomem *regs;
+ u32 num_rings;
+ struct flexrm_ring *rings;
+ u64 dma_mask;
+ struct dma_pool *bd_pool;
+ struct dma_pool *cmpl_pool;
+ struct mbox_controller controller;
+};
+
+static int flexrm_new_request(struct flexrm_ring *ring,
+ struct brcm_message *batch_msg,
+ struct brcm_message *msg)
+{
+ void *next;
+ unsigned long flags;
+ u32 val, count, nhcnt;
+ u32 read_offset, write_offset;
+ bool exit_cleanup = false;
+ int ret = 0, reqid;
+
+ /* Do sanity check on message */
+ if (!flexrm_sanity_check(msg))
+ return -EIO;
+ msg->error = 0;
+
+ /* If no requests possible then save data pointer and goto done. */
+ reqid = ida_simple_get(&ring->requests_ida, 0,
+ RING_MAX_REQ_COUNT, GFP_KERNEL);
+ if (reqid < 0) {
+ spin_lock_irqsave(&ring->lock, flags);
+ if (batch_msg)
+ ring->last_pending_msg = batch_msg;
+ else
+ ring->last_pending_msg = msg;
+ spin_unlock_irqrestore(&ring->lock, flags);
+ return 0;
+ }
+ ring->requests[reqid] = msg;
+
+ /* Do DMA mappings for the message */
+ ret = flexrm_dma_map(ring->mbox->dev, msg);
+ if (ret < 0) {
+ ring->requests[reqid] = NULL;
+ ida_simple_remove(&ring->requests_ida, reqid);
+ return ret;
+ }
+
+ /* If last_pending_msg is already set then goto done with error */
+ spin_lock_irqsave(&ring->lock, flags);
+ if (ring->last_pending_msg)
+ ret = -ENOSPC;
+ spin_unlock_irqrestore(&ring->lock, flags);
+ if (ret < 0) {
+ dev_warn(ring->mbox->dev, "no space in ring %d\n", ring->num);
+ exit_cleanup = true;
+ goto exit;
+ }
+
+ /* Determine current HW BD read offset */
+ read_offset = readl_relaxed(ring->regs + RING_BD_READ_PTR);
+ val = readl_relaxed(ring->regs + RING_BD_START_ADDR);
+ read_offset *= RING_DESC_SIZE;
+ read_offset += (u32)(BD_START_ADDR_DECODE(val) - ring->bd_dma_base);
+
+ /*
+ * Number required descriptors = number of non-header descriptors +
+ * number of header descriptors +
+ * 1x null descriptor
+ */
+ nhcnt = flexrm_estimate_nonheader_desc_count(msg);
+ count = flexrm_estimate_header_desc_count(nhcnt) + nhcnt + 1;
+
+ /* Check for available descriptor space. */
+ write_offset = ring->bd_write_offset;
+ while (count) {
+ if (!flexrm_is_next_table_desc(ring->bd_base + write_offset))
+ count--;
+ write_offset += RING_DESC_SIZE;
+ if (write_offset == RING_BD_SIZE)
+ write_offset = 0x0;
+ if (write_offset == read_offset)
+ break;
+ }
+ if (count) {
+ spin_lock_irqsave(&ring->lock, flags);
+ if (batch_msg)
+ ring->last_pending_msg = batch_msg;
+ else
+ ring->last_pending_msg = msg;
+ spin_unlock_irqrestore(&ring->lock, flags);
+ ret = 0;
+ exit_cleanup = true;
+ goto exit;
+ }
+
+ /* Write descriptors to ring */
+ next = flexrm_write_descs(msg, nhcnt, reqid,
+ ring->bd_base + ring->bd_write_offset,
+ RING_BD_TOGGLE_VALID(ring->bd_write_offset),
+ ring->bd_base, ring->bd_base + RING_BD_SIZE);
+ if (IS_ERR(next)) {
+ ret = PTR_ERR(next);
+ exit_cleanup = true;
+ goto exit;
+ }
+
+ /* Save ring BD write offset */
+ ring->bd_write_offset = (unsigned long)(next - ring->bd_base);
+
+exit:
+ /* Update error status in message */
+ msg->error = ret;
+
+ /* Cleanup if we failed */
+ if (exit_cleanup) {
+ flexrm_dma_unmap(ring->mbox->dev, msg);
+ ring->requests[reqid] = NULL;
+ ida_simple_remove(&ring->requests_ida, reqid);
+ }
+
+ return ret;
+}
+
+static int flexrm_process_completions(struct flexrm_ring *ring)
+{
+ u64 desc;
+ int err, count = 0;
+ unsigned long flags;
+ struct brcm_message *msg = NULL;
+ u32 reqid, cmpl_read_offset, cmpl_write_offset;
+ struct mbox_chan *chan = &ring->mbox->controller.chans[ring->num];
+
+ spin_lock_irqsave(&ring->lock, flags);
+
+ /* Check last_pending_msg */
+ if (ring->last_pending_msg) {
+ msg = ring->last_pending_msg;
+ ring->last_pending_msg = NULL;
+ }
+
+ /*
+ * Get current completion read and write offset
+ *
+ * Note: We should read completion write pointer atleast once
+ * after we get a MSI interrupt because HW maintains internal
+ * MSI status which will allow next MSI interrupt only after
+ * completion write pointer is read.
+ */
+ cmpl_write_offset = readl_relaxed(ring->regs + RING_CMPL_WRITE_PTR);
+ cmpl_write_offset *= RING_DESC_SIZE;
+ cmpl_read_offset = ring->cmpl_read_offset;
+ ring->cmpl_read_offset = cmpl_write_offset;
+
+ spin_unlock_irqrestore(&ring->lock, flags);
+
+ /* If last_pending_msg was set then queue it back */
+ if (msg)
+ mbox_send_message(chan, msg);
+
+ /* For each completed request notify mailbox clients */
+ reqid = 0;
+ while (cmpl_read_offset != cmpl_write_offset) {
+ /* Dequeue next completion descriptor */
+ desc = *((u64 *)(ring->cmpl_base + cmpl_read_offset));
+
+ /* Next read offset */
+ cmpl_read_offset += RING_DESC_SIZE;
+ if (cmpl_read_offset == RING_CMPL_SIZE)
+ cmpl_read_offset = 0;
+
+ /* Decode error from completion descriptor */
+ err = flexrm_cmpl_desc_to_error(desc);
+ if (err < 0) {
+ dev_warn(ring->mbox->dev,
+ "got completion desc=0x%lx with error %d",
+ (unsigned long)desc, err);
+ }
+
+ /* Determine request id from completion descriptor */
+ reqid = flexrm_cmpl_desc_to_reqid(desc);
+
+ /* Determine message pointer based on reqid */
+ msg = ring->requests[reqid];
+ if (!msg) {
+ dev_warn(ring->mbox->dev,
+ "null msg pointer for completion desc=0x%lx",
+ (unsigned long)desc);
+ continue;
+ }
+
+ /* Release reqid for recycling */
+ ring->requests[reqid] = NULL;
+ ida_simple_remove(&ring->requests_ida, reqid);
+
+ /* Unmap DMA mappings */
+ flexrm_dma_unmap(ring->mbox->dev, msg);
+
+ /* Give-back message to mailbox client */
+ msg->error = err;
+ mbox_chan_received_data(chan, msg);
+
+ /* Increment number of completions processed */
+ count++;
+ }
+
+ return count;
+}
+
+static irqreturn_t flexrm_irq_event(int irq, void *dev_id)
+{
+ /* We only have MSI for completions so just wakeup IRQ thread */
+ /* Ring related errors will be informed via completion descriptors */
+
+ return IRQ_WAKE_THREAD;
+}
+
+static irqreturn_t flexrm_irq_thread(int irq, void *dev_id)
+{
+ flexrm_process_completions(dev_id);
+
+ return IRQ_HANDLED;
+}
+
+static int flexrm_send_data(struct mbox_chan *chan, void *data)
+{
+ int i, rc;
+ struct flexrm_ring *ring = chan->con_priv;
+ struct brcm_message *msg = data;
+
+ if (msg->type == BRCM_MESSAGE_BATCH) {
+ for (i = msg->batch.msgs_queued;
+ i < msg->batch.msgs_count; i++) {
+ rc = flexrm_new_request(ring, msg,
+ &msg->batch.msgs[i]);
+ if (rc) {
+ msg->error = rc;
+ return rc;
+ }
+ msg->batch.msgs_queued++;
+ }
+ return 0;
+ }
+
+ return flexrm_new_request(ring, NULL, data);
+}
+
+static bool flexrm_peek_data(struct mbox_chan *chan)
+{
+ int cnt = flexrm_process_completions(chan->con_priv);
+
+ return (cnt > 0) ? true : false;
+}
+
+static int flexrm_startup(struct mbox_chan *chan)
+{
+ u64 d;
+ u32 val, off;
+ int ret = 0;
+ dma_addr_t next_addr;
+ struct flexrm_ring *ring = chan->con_priv;
+
+ /* Allocate BD memory */
+ ring->bd_base = dma_pool_alloc(ring->mbox->bd_pool,
+ GFP_KERNEL, &ring->bd_dma_base);
+ if (!ring->bd_base) {
+ dev_err(ring->mbox->dev, "can't allocate BD memory\n");
+ ret = -ENOMEM;
+ goto fail;
+ }
+
+ /* Configure next table pointer entries in BD memory */
+ for (off = 0; off < RING_BD_SIZE; off += RING_DESC_SIZE) {
+ next_addr = off + RING_DESC_SIZE;
+ if (next_addr == RING_BD_SIZE)
+ next_addr = 0;
+ next_addr += ring->bd_dma_base;
+ if (RING_BD_ALIGN_CHECK(next_addr))
+ d = flexrm_next_table_desc(RING_BD_TOGGLE_VALID(off),
+ next_addr);
+ else
+ d = flexrm_null_desc(RING_BD_TOGGLE_INVALID(off));
+ flexrm_write_desc(ring->bd_base + off, d);
+ }
+
+ /* Allocate completion memory */
+ ring->cmpl_base = dma_pool_alloc(ring->mbox->cmpl_pool,
+ GFP_KERNEL, &ring->cmpl_dma_base);
+ if (!ring->cmpl_base) {
+ dev_err(ring->mbox->dev, "can't allocate completion memory\n");
+ ret = -ENOMEM;
+ goto fail_free_bd_memory;
+ }
+ memset(ring->cmpl_base, 0, RING_CMPL_SIZE);
+
+ /* Request IRQ */
+ if (ring->irq == UINT_MAX) {
+ dev_err(ring->mbox->dev, "ring IRQ not available\n");
+ ret = -ENODEV;
+ goto fail_free_cmpl_memory;
+ }
+ ret = request_threaded_irq(ring->irq,
+ flexrm_irq_event,
+ flexrm_irq_thread,
+ 0, dev_name(ring->mbox->dev), ring);
+ if (ret) {
+ dev_err(ring->mbox->dev, "failed to request ring IRQ\n");
+ goto fail_free_cmpl_memory;
+ }
+ ring->irq_requested = true;
+
+ /* Disable/inactivate ring */
+ writel_relaxed(0x0, ring->regs + RING_CONTROL);
+
+ /* Program BD start address */
+ val = BD_START_ADDR_VALUE(ring->bd_dma_base);
+ writel_relaxed(val, ring->regs + RING_BD_START_ADDR);
+
+ /* BD write pointer will be same as HW write pointer */
+ ring->bd_write_offset =
+ readl_relaxed(ring->regs + RING_BD_WRITE_PTR);
+ ring->bd_write_offset *= RING_DESC_SIZE;
+
+ /* Program completion start address */
+ val = CMPL_START_ADDR_VALUE(ring->cmpl_dma_base);
+ writel_relaxed(val, ring->regs + RING_CMPL_START_ADDR);
+
+ /* Ensure last pending message is cleared */
+ ring->last_pending_msg = NULL;
+
+ /* Completion read pointer will be same as HW write pointer */
+ ring->cmpl_read_offset =
+ readl_relaxed(ring->regs + RING_CMPL_WRITE_PTR);
+ ring->cmpl_read_offset *= RING_DESC_SIZE;
+
+ /* Read ring Tx, Rx, and Outstanding counts to clear */
+ readl_relaxed(ring->regs + RING_NUM_REQ_RECV_LS);
+ readl_relaxed(ring->regs + RING_NUM_REQ_RECV_MS);
+ readl_relaxed(ring->regs + RING_NUM_REQ_TRANS_LS);
+ readl_relaxed(ring->regs + RING_NUM_REQ_TRANS_MS);
+ readl_relaxed(ring->regs + RING_NUM_REQ_OUTSTAND);
+
+ /* Configure RING_MSI_CONTROL */
+ val = 0;
+ val |= (ring->msi_timer_val << MSI_TIMER_VAL_SHIFT);
+ val |= BIT(MSI_ENABLE_SHIFT);
+ val |= (ring->msi_count_threshold & MSI_COUNT_MASK) << MSI_COUNT_SHIFT;
+ writel_relaxed(val, ring->regs + RING_MSI_CONTROL);
+
+ /* Enable/activate ring */
+ val = BIT(CONTROL_ACTIVE_SHIFT);
+ writel_relaxed(val, ring->regs + RING_CONTROL);
+
+ return 0;
+
+fail_free_cmpl_memory:
+ dma_pool_free(ring->mbox->cmpl_pool,
+ ring->cmpl_base, ring->cmpl_dma_base);
+ ring->cmpl_base = NULL;
+fail_free_bd_memory:
+ dma_pool_free(ring->mbox->bd_pool,
+ ring->bd_base, ring->bd_dma_base);
+ ring->bd_base = NULL;
+fail:
+ return ret;
+}
+
+static void flexrm_shutdown(struct mbox_chan *chan)
+{
+ u32 reqid;
+ unsigned int timeout;
+ struct brcm_message *msg;
+ struct flexrm_ring *ring = chan->con_priv;
+
+ /* Disable/inactivate ring */
+ writel_relaxed(0x0, ring->regs + RING_CONTROL);
+
+ /* Flush ring with timeout of 1s */
+ timeout = 1000;
+ writel_relaxed(BIT(CONTROL_FLUSH_SHIFT),
+ ring->regs + RING_CONTROL);
+ do {
+ if (readl_relaxed(ring->regs + RING_FLUSH_DONE) &
+ FLUSH_DONE_MASK)
+ break;
+ mdelay(1);
+ } while (timeout--);
+
+ /* Abort all in-flight requests */
+ for (reqid = 0; reqid < RING_MAX_REQ_COUNT; reqid++) {
+ msg = ring->requests[reqid];
+ if (!msg)
+ continue;
+
+ /* Release reqid for recycling */
+ ring->requests[reqid] = NULL;
+ ida_simple_remove(&ring->requests_ida, reqid);
+
+ /* Unmap DMA mappings */
+ flexrm_dma_unmap(ring->mbox->dev, msg);
+
+ /* Give-back message to mailbox client */
+ msg->error = -EIO;
+ mbox_chan_received_data(chan, msg);
+ }
+
+ /* Release IRQ */
+ if (ring->irq_requested) {
+ free_irq(ring->irq, ring);
+ ring->irq_requested = false;
+ }
+
+ /* Free-up completion descriptor ring */
+ if (ring->cmpl_base) {
+ dma_pool_free(ring->mbox->cmpl_pool,
+ ring->cmpl_base, ring->cmpl_dma_base);
+ ring->cmpl_base = NULL;
+ }
+
+ /* Free-up BD descriptor ring */
+ if (ring->bd_base) {
+ dma_pool_free(ring->mbox->bd_pool,
+ ring->bd_base, ring->bd_dma_base);
+ ring->bd_base = NULL;
+ }
+}
+
+static bool flexrm_last_tx_done(struct mbox_chan *chan)
+{
+ bool ret;
+ unsigned long flags;
+ struct flexrm_ring *ring = chan->con_priv;
+
+ spin_lock_irqsave(&ring->lock, flags);
+ ret = (ring->last_pending_msg) ? false : true;
+ spin_unlock_irqrestore(&ring->lock, flags);
+
+ return ret;
+}
+
+static const struct mbox_chan_ops flexrm_mbox_chan_ops = {
+ .send_data = flexrm_send_data,
+ .startup = flexrm_startup,
+ .shutdown = flexrm_shutdown,
+ .last_tx_done = flexrm_last_tx_done,
+ .peek_data = flexrm_peek_data,
+};
+
+static void flexrm_mbox_msi_write(struct msi_desc *desc, struct msi_msg *msg)
+{
+ struct device *dev = msi_desc_to_dev(desc);
+ struct flexrm_mbox *mbox = dev_get_drvdata(dev);
+ struct flexrm_ring *ring = &mbox->rings[desc->platform.msi_index];
+
+ /* Configure per-Ring MSI registers */
+ writel_relaxed(msg->address_lo, ring->regs + RING_MSI_ADDR_LS);
+ writel_relaxed(msg->address_hi, ring->regs + RING_MSI_ADDR_MS);
+ writel_relaxed(msg->data, ring->regs + RING_MSI_DATA_VALUE);
+}
+
+static struct mbox_chan *flexrm_mbox_of_xlate(struct mbox_controller *cntlr,
+ const struct of_phandle_args *pa)
+{
+ struct mbox_chan *chan;
+ struct flexrm_ring *ring;
+
+ if (pa->args_count < 3)
+ return ERR_PTR(-EINVAL);
+
+ if (pa->args[0] >= cntlr->num_chans)
+ return ERR_PTR(-ENOENT);
+
+ if (pa->args[1] > MSI_COUNT_MASK)
+ return ERR_PTR(-EINVAL);
+
+ if (pa->args[2] > MSI_TIMER_VAL_MASK)
+ return ERR_PTR(-EINVAL);
+
+ chan = &cntlr->chans[pa->args[0]];
+ ring = chan->con_priv;
+ ring->msi_count_threshold = pa->args[1];
+ ring->msi_timer_val = pa->args[2];
+
+ return chan;
+}
+
+static int flexrm_mbox_probe(struct platform_device *pdev)
+{
+ int index, ret = 0;
+ void __iomem *regs;
+ void __iomem *regs_end;
+ struct msi_desc *desc;
+ struct resource *iomem;
+ struct flexrm_ring *ring;
+ struct flexrm_mbox *mbox;
+ struct device *dev = &pdev->dev;
+
+ /* Allocate driver mailbox struct */
+ mbox = devm_kzalloc(dev, sizeof(*mbox), GFP_KERNEL);
+ if (!mbox) {
+ ret = -ENOMEM;
+ goto fail;
+ }
+ mbox->dev = dev;
+ platform_set_drvdata(pdev, mbox);
+
+ /* Get resource for registers */
+ iomem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ if (!iomem || (resource_size(iomem) < RING_REGS_SIZE)) {
+ ret = -ENODEV;
+ goto fail;
+ }
+
+ /* Map registers of all rings */
+ mbox->regs = devm_ioremap_resource(&pdev->dev, iomem);
+ if (IS_ERR(mbox->regs)) {
+ ret = PTR_ERR(mbox->regs);
+ dev_err(&pdev->dev, "Failed to remap mailbox regs: %d\n", ret);
+ goto fail;
+ }
+ regs_end = mbox->regs + resource_size(iomem);
+
+ /* Scan and count available rings */
+ mbox->num_rings = 0;
+ for (regs = mbox->regs; regs < regs_end; regs += RING_REGS_SIZE) {
+ if (readl_relaxed(regs + RING_VER) == RING_VER_MAGIC)
+ mbox->num_rings++;
+ }
+ if (!mbox->num_rings) {
+ ret = -ENODEV;
+ goto fail;
+ }
+
+ /* Allocate driver ring structs */
+ ring = devm_kcalloc(dev, mbox->num_rings, sizeof(*ring), GFP_KERNEL);
+ if (!ring) {
+ ret = -ENOMEM;
+ goto fail;
+ }
+ mbox->rings = ring;
+
+ /* Initialize members of driver ring structs */
+ regs = mbox->regs;
+ for (index = 0; index < mbox->num_rings; index++) {
+ ring = &mbox->rings[index];
+ ring->num = index;
+ ring->mbox = mbox;
+ while ((regs < regs_end) &&
+ (readl_relaxed(regs + RING_VER) != RING_VER_MAGIC))
+ regs += RING_REGS_SIZE;
+ if (regs_end <= regs) {
+ ret = -ENODEV;
+ goto fail;
+ }
+ ring->regs = regs;
+ regs += RING_REGS_SIZE;
+ ring->irq = UINT_MAX;
+ ring->irq_requested = false;
+ ring->msi_timer_val = MSI_TIMER_VAL_MASK;
+ ring->msi_count_threshold = 0x1;
+ ida_init(&ring->requests_ida);
+ memset(ring->requests, 0, sizeof(ring->requests));
+ ring->bd_base = NULL;
+ ring->bd_dma_base = 0;
+ ring->cmpl_base = NULL;
+ ring->cmpl_dma_base = 0;
+ spin_lock_init(&ring->lock);
+ ring->last_pending_msg = NULL;
+ ring->cmpl_read_offset = 0;
+ }
+
+ /* FlexRM is capable of 40-bit physical addresses only */
+ mbox->dma_mask = DMA_BIT_MASK(40);
+ dev->dma_mask = &mbox->dma_mask;
+
+ /* Create DMA pool for ring BD memory */
+ mbox->bd_pool = dma_pool_create("bd", dev, RING_BD_SIZE,
+ 1 << RING_BD_ALIGN_ORDER, 0);
+ if (!mbox->bd_pool) {
+ ret = -ENOMEM;
+ goto fail;
+ }
+
+ /* Create DMA pool for ring completion memory */
+ mbox->cmpl_pool = dma_pool_create("cmpl", dev, RING_CMPL_SIZE,
+ 1 << RING_CMPL_ALIGN_ORDER, 0);
+ if (!mbox->cmpl_pool) {
+ ret = -ENOMEM;
+ goto fail_destroy_bd_pool;
+ }
+
+ /* Allocate platform MSIs for each ring */
+ ret = platform_msi_domain_alloc_irqs(dev, mbox->num_rings,
+ flexrm_mbox_msi_write);
+ if (ret)
+ goto fail_destroy_cmpl_pool;
+
+ /* Save alloced IRQ numbers for each ring */
+ for_each_msi_entry(desc, dev) {
+ ring = &mbox->rings[desc->platform.msi_index];
+ ring->irq = desc->irq;
+ }
+
+ /* Initialize mailbox controller */
+ mbox->controller.txdone_irq = false;
+ mbox->controller.txdone_poll = true;
+ mbox->controller.txpoll_period = 1;
+ mbox->controller.ops = &flexrm_mbox_chan_ops;
+ mbox->controller.dev = dev;
+ mbox->controller.num_chans = mbox->num_rings;
+ mbox->controller.of_xlate = flexrm_mbox_of_xlate;
+ mbox->controller.chans = devm_kcalloc(dev, mbox->num_rings,
+ sizeof(*mbox->controller.chans), GFP_KERNEL);
+ if (!mbox->controller.chans) {
+ ret = -ENOMEM;
+ goto fail_free_msis;
+ }
+ for (index = 0; index < mbox->num_rings; index++)
+ mbox->controller.chans[index].con_priv = &mbox->rings[index];
+
+ /* Register mailbox controller */
+ ret = mbox_controller_register(&mbox->controller);
+ if (ret)
+ goto fail_free_msis;
+
+ dev_info(dev, "registered flexrm mailbox with %d channels\n",
+ mbox->controller.num_chans);
+
+ return 0;
+
+fail_free_msis:
+ platform_msi_domain_free_irqs(dev);
+fail_destroy_cmpl_pool:
+ dma_pool_destroy(mbox->cmpl_pool);
+fail_destroy_bd_pool:
+ dma_pool_destroy(mbox->bd_pool);
+fail:
+ return ret;
+}
+
+static int flexrm_mbox_remove(struct platform_device *pdev)
+{
+ int index;
+ struct device *dev = &pdev->dev;
+ struct flexrm_ring *ring;
+ struct flexrm_mbox *mbox = platform_get_drvdata(pdev);
+
+ mbox_controller_unregister(&mbox->controller);
+
+ platform_msi_domain_free_irqs(dev);
+
+ dma_pool_destroy(mbox->cmpl_pool);
+ dma_pool_destroy(mbox->bd_pool);
+
+ for (index = 0; index < mbox->num_rings; index++) {
+ ring = &mbox->rings[index];
+ ida_destroy(&ring->requests_ida);
+ }
+
+ return 0;
+}
+
+static const struct of_device_id flexrm_mbox_of_match[] = {
+ { .compatible = "brcm,iproc-flexrm-mbox", },
+ {},
+};
+MODULE_DEVICE_TABLE(of, flexrm_mbox_of_match);
+
+static struct platform_driver flexrm_mbox_driver = {
+ .driver = {
+ .name = "brcm-flexrm-mbox",
+ .of_match_table = flexrm_mbox_of_match,
+ },
+ .probe = flexrm_mbox_probe,
+ .remove = flexrm_mbox_remove,
+};
+module_platform_driver(flexrm_mbox_driver);
+
+MODULE_AUTHOR("Anup Patel <anup.patel@broadcom.com>");
+MODULE_DESCRIPTION("Broadcom FlexRM mailbox driver");
+MODULE_LICENSE("GPL v2");
diff --git a/include/linux/mailbox/brcm-message.h b/include/linux/mailbox/brcm-message.h
index 6b55c93..c20b484 100644
--- a/include/linux/mailbox/brcm-message.h
+++ b/include/linux/mailbox/brcm-message.h
@@ -16,6 +16,7 @@
enum brcm_message_type {
BRCM_MESSAGE_UNKNOWN = 0,
+ BRCM_MESSAGE_BATCH,
BRCM_MESSAGE_SPU,
BRCM_MESSAGE_SBA,
BRCM_MESSAGE_MAX,
@@ -23,24 +24,29 @@ enum brcm_message_type {
struct brcm_sba_command {
u64 cmd;
+ u64 *cmd_dma;
+ dma_addr_t cmd_dma_addr;
#define BRCM_SBA_CMD_TYPE_A BIT(0)
#define BRCM_SBA_CMD_TYPE_B BIT(1)
#define BRCM_SBA_CMD_TYPE_C BIT(2)
#define BRCM_SBA_CMD_HAS_RESP BIT(3)
#define BRCM_SBA_CMD_HAS_OUTPUT BIT(4)
u64 flags;
- dma_addr_t input;
- size_t input_len;
dma_addr_t resp;
size_t resp_len;
- dma_addr_t output;
- size_t output_len;
+ dma_addr_t data;
+ size_t data_len;
};
struct brcm_message {
enum brcm_message_type type;
union {
struct {
+ struct brcm_message *msgs;
+ unsigned int msgs_queued;
+ unsigned int msgs_count;
+ } batch;
+ struct {
struct scatterlist *src;
struct scatterlist *dst;
} spu;
--
2.7.4
^ permalink raw reply related
* [PATCH v3 2/2] dt-bindings: Add DT bindings info for FlexRM ring manager
From: Anup Patel @ 2017-01-04 5:34 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483508082-7008-1-git-send-email-anup.patel@broadcom.com>
This patch adds device tree bindings document for the FlexRM
ring manager found on Broadcom iProc SoCs.
Reviewed-by: Ray Jui <ray.jui@broadcom.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
Signed-off-by: Anup Patel <anup.patel@broadcom.com>
---
.../bindings/mailbox/brcm,iproc-flexrm-mbox.txt | 60 ++++++++++++++++++++++
1 file changed, 60 insertions(+)
create mode 100644 Documentation/devicetree/bindings/mailbox/brcm,iproc-flexrm-mbox.txt
diff --git a/Documentation/devicetree/bindings/mailbox/brcm,iproc-flexrm-mbox.txt b/Documentation/devicetree/bindings/mailbox/brcm,iproc-flexrm-mbox.txt
new file mode 100644
index 0000000..ca51a39
--- /dev/null
+++ b/Documentation/devicetree/bindings/mailbox/brcm,iproc-flexrm-mbox.txt
@@ -0,0 +1,60 @@
+Broadcom FlexRM Ring Manager
+============================
+The Broadcom FlexRM ring manager provides a set of rings which can be
+used to submit work to offload engines. An SoC may have multiple FlexRM
+hardware blocks. There is one device tree entry per FlexRM block. The
+FlexRM driver will create a mailbox-controller instance for given FlexRM
+hardware block where each mailbox channel is a separate FlexRM ring.
+
+Required properties:
+--------------------
+- compatible: Should be "brcm,iproc-flexrm-mbox"
+- reg: Specifies base physical address and size of the FlexRM
+ ring registers
+- msi-parent: Phandles (and potential Device IDs) to MSI controllers
+ The FlexRM engine will send MSIs (instead of wired
+ interrupts) to CPU. There is one MSI for each FlexRM ring.
+ Refer devicetree/bindings/interrupt-controller/msi.txt
+- #mbox-cells: Specifies the number of cells needed to encode a mailbox
+ channel. This should be 3.
+
+ The 1st cell is the mailbox channel number.
+
+ The 2nd cell contains MSI completion threshold. This is the
+ number of completion messages for which FlexRM will inject
+ one MSI interrupt to CPU.
+
+ The 3nd cell contains MSI timer value representing time for
+ which FlexRM will wait to accumulate N completion messages
+ where N is the value specified by 2nd cell above. If FlexRM
+ does not get required number of completion messages in time
+ specified by this cell then it will inject one MSI interrupt
+ to CPU provided atleast one completion message is available.
+
+Optional properties:
+--------------------
+- dma-coherent: Present if DMA operations made by the FlexRM engine (such
+ as DMA descriptor access, access to buffers pointed by DMA
+ descriptors and read/write pointer updates to DDR) are
+ cache coherent with the CPU.
+
+Example:
+--------
+crypto_mbox: mbox at 67000000 {
+ compatible = "brcm,iproc-flexrm-mbox";
+ reg = <0x67000000 0x200000>;
+ msi-parent = <&gic_its 0x7f00>;
+ #mbox-cells = <3>;
+};
+
+crypto_client {
+ ...
+ mboxes = <&crypto_mbox 0 0x1 0xffff>,
+ <&crypto_mbox 1 0x1 0xffff>,
+ <&crypto_mbox 16 0x1 0xffff>,
+ <&crypto_mbox 17 0x1 0xffff>,
+ <&crypto_mbox 30 0x1 0xffff>,
+ <&crypto_mbox 31 0x1 0xffff>;
+ };
+ ...
+};
--
2.7.4
^ permalink raw reply related
* [GIT PULL] Amlogic fixes for v4.10-rc
From: Kevin Hilman @ 2017-01-04 5:35 UTC (permalink / raw)
To: linux-arm-kernel
Arnd, Olof,
This pull has one real fix, as a couple non-critical ones. The DRM
DT/defconfig patches are coming now because I didn't expect the new
driver to make it for the v4.10 merge window, but since it did[1], the
DT and defconfig should go into the same release.
Thanks,
Kevin
[1] bbbe775ec5b5 drm: Add support for Amlogic Meson Graphic Controller
The following changes since commit 7ce7d89f48834cefece7804d38fc5d85382edf77:
Linux 4.10-rc1 (2016-12-25 16:13:08 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-amlogic.git tags/amlogic-fixes
for you to fetch changes up to fcdaf1a2a7a042a290f4c7de28bcdebd5de18445:
ARM64: defconfig: enable DRM_MESON as module (2017-01-03 09:31:13 -0800)
----------------------------------------------------------------
Amlogic fixes for v4.10
- DT: GXL: fix GPIO include
- add DT and defconfig for newly merged DRM driver
----------------------------------------------------------------
Kevin Hilman (2):
ARM64: dts: meson-gxl: fix GPIO include
ARM64: defconfig: enable DRM_MESON as module
Neil Armstrong (1):
ARM64: dts: meson-gx: Add Graphic Controller nodes
arch/arm64/boot/dts/amlogic/meson-gx.dtsi | 16 ++++++++++++++++
arch/arm64/boot/dts/amlogic/meson-gxbb-nexbox-a95x.dts | 16 ++++++++++++++++
arch/arm64/boot/dts/amlogic/meson-gxbb-p20x.dtsi | 16 ++++++++++++++++
arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi | 4 ++++
arch/arm64/boot/dts/amlogic/meson-gxl-nexbox-a95x.dts | 16 ++++++++++++++++
arch/arm64/boot/dts/amlogic/meson-gxl.dtsi | 6 +++++-
arch/arm64/boot/dts/amlogic/meson-gxm-nexbox-a1.dts | 16 ++++++++++++++++
arch/arm64/boot/dts/amlogic/meson-gxm.dtsi | 4 ++++
arch/arm64/configs/defconfig | 1 +
9 files changed, 94 insertions(+), 1 deletion(-)
^ permalink raw reply
* [PATCH v5 2/4] soc: zte: pm_domains: Prepare for supporting ARMv8 zx2967 family
From: Shawn Guo @ 2017-01-04 5:44 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483489157-10782-2-git-send-email-baoyou.xie@linaro.org>
On Wed, Jan 04, 2017 at 08:19:15AM +0800, Baoyou Xie wrote:
> The ARMv8 zx2967 family (296718, 296716 etc) uses different value
> for controlling the power domain on/off registers, Choose the
> value depending on the compatible.
>
> Multiple domains are prepared for the family, this patch prepares
> the common functions.
>
> Signed-off-by: Baoyou Xie <baoyou.xie@linaro.org>
> Reviewed-by: Shawn Guo <shawnguo@kernel.org>
> Reviewed-by: Jun Nie <jun.nie@linaro.org>
Same here. Neither myself nor Jun gave Reviewed-by tag on this patch.
> ---
> drivers/soc/Kconfig | 1 +
> drivers/soc/Makefile | 1 +
> drivers/soc/zte/Kconfig | 13 ++++
> drivers/soc/zte/Makefile | 4 ++
> drivers/soc/zte/zx2967_pm_domains.c | 139 ++++++++++++++++++++++++++++++++++++
> drivers/soc/zte/zx2967_pm_domains.h | 46 ++++++++++++
> 6 files changed, 204 insertions(+)
> create mode 100644 drivers/soc/zte/Kconfig
> create mode 100644 drivers/soc/zte/Makefile
> create mode 100644 drivers/soc/zte/zx2967_pm_domains.c
> create mode 100644 drivers/soc/zte/zx2967_pm_domains.h
>
> diff --git a/drivers/soc/Kconfig b/drivers/soc/Kconfig
> index f31bceb..f09023f 100644
> --- a/drivers/soc/Kconfig
> +++ b/drivers/soc/Kconfig
> @@ -11,5 +11,6 @@ source "drivers/soc/tegra/Kconfig"
> source "drivers/soc/ti/Kconfig"
> source "drivers/soc/ux500/Kconfig"
> source "drivers/soc/versatile/Kconfig"
> +source "drivers/soc/zte/Kconfig"
>
> endmenu
> diff --git a/drivers/soc/Makefile b/drivers/soc/Makefile
> index 50c23d0..05eae52 100644
> --- a/drivers/soc/Makefile
> +++ b/drivers/soc/Makefile
> @@ -16,3 +16,4 @@ obj-$(CONFIG_ARCH_TEGRA) += tegra/
> obj-$(CONFIG_SOC_TI) += ti/
> obj-$(CONFIG_ARCH_U8500) += ux500/
> obj-$(CONFIG_PLAT_VERSATILE) += versatile/
> +obj-$(CONFIG_ARCH_ZX) += zte/
> diff --git a/drivers/soc/zte/Kconfig b/drivers/soc/zte/Kconfig
> new file mode 100644
> index 0000000..20bde38
> --- /dev/null
> +++ b/drivers/soc/zte/Kconfig
> @@ -0,0 +1,13 @@
> +#
> +# ZTE SoC drivers
> +#
> +menuconfig SOC_ZTE
> + bool "ZTE SoC driver support"
> +
> +if SOC_ZTE
> +
> +config ZX2967_PM_DOMAINS
> + bool "ZX2967 PM domains"
> + depends on PM_GENERIC_DOMAINS
> +
> +endif
> diff --git a/drivers/soc/zte/Makefile b/drivers/soc/zte/Makefile
> new file mode 100644
> index 0000000..8a37f2f
> --- /dev/null
> +++ b/drivers/soc/zte/Makefile
> @@ -0,0 +1,4 @@
> +#
> +# ZTE SOC drivers
> +#
> +obj-$(CONFIG_ZX2967_PM_DOMAINS) += zx2967_pm_domains.o
> diff --git a/drivers/soc/zte/zx2967_pm_domains.c b/drivers/soc/zte/zx2967_pm_domains.c
> new file mode 100644
> index 0000000..a215875
> --- /dev/null
> +++ b/drivers/soc/zte/zx2967_pm_domains.c
> @@ -0,0 +1,139 @@
> +/*
> + * Copyright (C) 2017 ZTE Ltd.
> + *
> + * Author: Baoyou Xie <baoyou.xie@linaro.org>
> + * License terms: GNU General Public License (GPL) version 2
> + */
> +#include <linux/delay.h>
> +#include <linux/err.h>
> +#include <linux/io.h>
> +#include <linux/of.h>
> +
> +#include "zx2967_pm_domains.h"
> +
> +#define PCU_DM_CLKEN(zpd) ((zpd)->reg_offset[REG_CLKEN])
> +#define PCU_DM_ISOEN(zpd) ((zpd)->reg_offset[REG_ISOEN])
> +#define PCU_DM_RSTEN(zpd) ((zpd)->reg_offset[REG_RSTEN])
> +#define PCU_DM_PWREN(zpd) ((zpd)->reg_offset[REG_PWREN])
> +#define PCU_DM_PWRDN(zpd) ((zpd)->reg_offset[REG_PWRDN])
PCU_DM_PWRDN doesn't seem to be used anywhere in this driver.
> +#define PCU_DM_ACK_SYNC(zpd) ((zpd)->reg_offset[REG_ACK_SYNC])
> +
> +static void __iomem *pcubase;
> +
> +int zx2967_power_on(struct generic_pm_domain *domain)
> +{
> + struct zx2967_pm_domain *zpd = (struct zx2967_pm_domain *)domain;
> + unsigned long loop = 1000;
> + u32 val;
> +
> + val = readl_relaxed(pcubase + PCU_DM_PWREN(zpd));
> + if (zpd->polarity == PWREN)
> + val |= BIT(zpd->bit);
> + else
> + val &= ~BIT(zpd->bit);
> + writel_relaxed(val, pcubase + PCU_DM_PWREN(zpd));
> +
> + do {
> + udelay(1);
> + val = readl_relaxed(pcubase + PCU_DM_ACK_SYNC(zpd))
> + & BIT(zpd->bit);
> + } while (--loop && !val);
> +
> + if (!loop) {
> + pr_err("Error: %s %s fail\n", __func__, domain->name);
> + return -EIO;
> + }
> +
> + val = readl_relaxed(pcubase + PCU_DM_RSTEN(zpd));
> + val |= BIT(zpd->bit);
> + writel_relaxed(val, pcubase + PCU_DM_RSTEN(zpd));
> + udelay(5);
> +
> + val = readl_relaxed(pcubase + PCU_DM_ISOEN(zpd));
> + val &= ~BIT(zpd->bit);
> + writel_relaxed(val, pcubase + PCU_DM_ISOEN(zpd));
> + udelay(5);
> +
> + val = readl_relaxed(pcubase + PCU_DM_CLKEN(zpd));
> + val |= BIT(zpd->bit);
> + writel_relaxed(val, pcubase + PCU_DM_CLKEN(zpd));
> + udelay(5);
> +
> + pr_debug("normal poweron %s\n", domain->name);
Does "normal" in the debug message mean anything here?
> +
> + return 0;
> +}
> +
> +int zx2967_power_off(struct generic_pm_domain *domain)
> +{
> + struct zx2967_pm_domain *zpd = (struct zx2967_pm_domain *)domain;
> + unsigned long loop = 1000;
> + u32 val;
> +
> + val = readl_relaxed(pcubase + PCU_DM_CLKEN(zpd));
> + val &= ~BIT(zpd->bit);
> + writel_relaxed(val, pcubase + PCU_DM_CLKEN(zpd));
> + udelay(5);
> +
> + val = readl_relaxed(pcubase + PCU_DM_ISOEN(zpd));
> + val |= BIT(zpd->bit);
> + writel_relaxed(val, pcubase + PCU_DM_ISOEN(zpd));
> + udelay(5);
> +
> + val = readl_relaxed(pcubase + PCU_DM_RSTEN(zpd));
> + val &= ~BIT(zpd->bit);
> + writel_relaxed(val, pcubase + PCU_DM_RSTEN(zpd));
> + udelay(5);
> +
> + val = readl_relaxed(pcubase + PCU_DM_PWREN(zpd));
> + if (zpd->polarity == PWREN)
> + val &= ~BIT(zpd->bit);
> + else
> + val |= BIT(zpd->bit);
> + writel_relaxed(val, pcubase + PCU_DM_PWREN(zpd));
> +
> + do {
> + udelay(1);
> + val = readl_relaxed(pcubase + PCU_DM_ACK_SYNC(zpd))
> + & BIT(zpd->bit);
> + } while (--loop && val);
> +
> + if (!loop) {
> + pr_err("Error: %s %s fail\n", __func__, domain->name);
> + return -EIO;
> + }
> +
> + pr_debug("normal poweroff %s\n", domain->name);
Ditto
> +
> + return 0;
> +}
> +
> +int zx2967_pd_probe(struct platform_device *pdev,
> + struct generic_pm_domain **zx_pm_domains,
> + int domain_num)
> +{
> + struct genpd_onecell_data *genpd_data;
> + struct resource *res;
> + int i;
> +
> + genpd_data = devm_kzalloc(&pdev->dev, sizeof(*genpd_data), GFP_KERNEL);
> + if (!genpd_data)
> + return -ENOMEM;
> +
> + genpd_data->domains = zx_pm_domains;
> + genpd_data->num_domains = domain_num;
> +
> + res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> + pcubase = devm_ioremap_resource(&pdev->dev, res);
> + if (IS_ERR(pcubase)) {
> + dev_err(&pdev->dev, "ioremap fail.\n");
> + return PTR_ERR(pcubase);
> + }
> +
> + for (i = 0; i < domain_num; ++i)
> + pm_genpd_init(zx_pm_domains[i], NULL, false);
> +
> + of_genpd_add_provider_onecell(pdev->dev.of_node, genpd_data);
> + dev_info(&pdev->dev, "powerdomain init ok\n");
> + return 0;
> +}
> diff --git a/drivers/soc/zte/zx2967_pm_domains.h b/drivers/soc/zte/zx2967_pm_domains.h
> new file mode 100644
> index 0000000..81ad4d6
> --- /dev/null
> +++ b/drivers/soc/zte/zx2967_pm_domains.h
> @@ -0,0 +1,46 @@
> +/*
> + * Header for ZTE's Power Domain Driver support
> + *
> + * Copyright (C) 2017 ZTE Ltd.
> + *
> + * Author: Baoyou Xie <baoyou.xie@linaro.org>
> + * License terms: GNU General Public License (GPL) version 2
> + */
> +
> +#ifndef __ZTE_ZX2967_PM_DOMAIN_H
> +#define __ZTE_ZX2967_PM_DOMAIN_H
> +
> +#include <linux/platform_device.h>
> +#include <linux/pm_domain.h>
> +
> +enum {
> + REG_CLKEN,
> + REG_ISOEN,
> + REG_RSTEN,
> + REG_PWREN,
> + REG_PWRDN,
> + REG_ACK_SYNC,
> +
> + /* The size of the array - must be last */
> + REG_ARRAY_SIZE,
> +};
> +
> +enum zx2967_power_polarity {
> + PWREN,
> + PWRDN,
> +};
> +
> +struct zx2967_pm_domain {
> + struct generic_pm_domain dm;
> + const u16 bit;
> + const enum zx2967_power_polarity polarity;
> + const u16 *reg_offset;
> +};
> +
> +extern int zx2967_power_on(struct generic_pm_domain *domain);
> +extern int zx2967_power_off(struct generic_pm_domain *domain);
> +extern int zx2967_pd_probe(struct platform_device *pdev,
> + struct generic_pm_domain **zx_pm_domains,
> + int domain_num);
The 'extern' is not really necessary here.
Shawn
> +
> +#endif /* __ZTE_ZX2967_PM_DOMAIN_H */
> --
> 2.7.4
>
^ permalink raw reply
* arm64: virt_to_page() does not return right page for a kernel image address
From: Pratyush Anand @ 2017-01-04 5:49 UTC (permalink / raw)
To: linux-arm-kernel
Hi,
I noticed that on arm64 kmap_atomic() does not return correct address
corresponding to a page located in data section. It causes crash in
kdump kernel with v29 kdump patches. crash happens in a newly
implemented crypto test [1], and the same test fails(even though it
does not crash) in 1st kernel as well.
Further debugging showed that the physical address returned by
virt_to_phys(kaddr) and virt_to_phys(kmap_atomic(virt_to_page(kaddr))
+ offset_in_page(kaddr)) are not same.
Mark Rutland thinks(IRC :#armlinux) that _virt_to_pgoff *only* handles
linear addresses, and not kernel image addresses. However, we have to
ask if it should?
Meanwhile, I reverted commit [2] and then everything worked fine
*atleast* in my case. But, I am not sure if that could be the right
and best solution.
Opinion?
~Pratyush
[1]
commit d7db7a882debaffc78f91aabedee973aa1f73390
Author: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Date: Fri Oct 21 13:19:54 2016 +0100
crypto: acomp - update testmgr with support for acomp
[2]commit 9f2875912dac35d9272a82ea9eec9e5884b42cd2
Author: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Date: Wed Mar 30 16:46:01 2016 +0200
arm64: mm: restrict virt_to_page() to the linear mapping
^ permalink raw reply
* [PATCH v5 3/4] soc: zte: pm_domains: Add support for zx296718 board
From: Shawn Guo @ 2017-01-04 5:49 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1483489157-10782-3-git-send-email-baoyou.xie@linaro.org>
On Wed, Jan 04, 2017 at 08:19:16AM +0800, Baoyou Xie wrote:
> This patch introduces the power domain driver of zx296718
> which belongs to zte's zx2967 family.
>
> Signed-off-by: Baoyou Xie <baoyou.xie@linaro.org>
> Reviewed-by: Shawn Guo <shawnguo@kernel.org>
> Reviewed-by: Jun Nie <jun.nie@linaro.org>
Jun did give his Reviewed-by tag on v2 of this patch, but I did not.
Once again, I put quite a few comments on v3 of this patch [1]. But
neither you responded to nor address any of them in reposting.
Shawn
[1] http://www.spinics.net/lists/arm-kernel/msg547691.html
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox