2010-06-10 02:27:11

by Dmytro Milinevskyy

[permalink] [raw]
Subject: [PATCH] Marvell OpenRD-Ultimate machine support

Hi!
This patch adds support for the OpenRD Ultimate machine(could be found at http://www.arm.linux.org.uk/developer/machines/list.php?id=2884)

Besides adding machine description this patch adds correction for PHY address for Ultimate version.

Differences from the previous attempt:
- Correctly filled instances of mv643xx_eth_platform_data in case of ultimate version.
- Do PCIE initialization w/o additional checks since all openrd types have it
- Init ge01 if (!openrd-base) to make the code cleaner

Regards,

-- Dmytro Milinevskyy

Signed-off-by: Dmytro Milinevskyy <[email protected]>
---
arch/arm/mach-kirkwood/Kconfig | 7 +++++++
arch/arm/mach-kirkwood/openrd-setup.c | 26 ++++++++++++++++++++++----
2 files changed, 29 insertions(+), 4 deletions(-)

diff --git a/arch/arm/mach-kirkwood/Kconfig b/arch/arm/mach-kirkwood/Kconfig
index 29b2163..110a3b8 100644
--- a/arch/arm/mach-kirkwood/Kconfig
+++ b/arch/arm/mach-kirkwood/Kconfig
@@ -75,6 +75,13 @@ config MACH_OPENRD_CLIENT
Say 'Y' here if you want your kernel to support the
Marvell OpenRD Client Board.

+config MACH_OPENRD_ULTIMATE
+ bool "Marvell OpenRD Ultimate Board"
+ select MACH_OPENRD
+ help
+ Say 'Y' here if you want your kernel to support the
+ Marvell OpenRD Ultimate Board.
+
config MACH_NETSPACE_V2
bool "LaCie Network Space v2 NAS Board"
help
diff --git a/arch/arm/mach-kirkwood/openrd-setup.c b/arch/arm/mach-kirkwood/openrd-setup.c
index ad3f1ec..c3d8c60 100644
--- a/arch/arm/mach-kirkwood/openrd-setup.c
+++ b/arch/arm/mach-kirkwood/openrd-setup.c
@@ -1,7 +1,7 @@
/*
* arch/arm/mach-kirkwood/openrd-setup.c
*
- * Marvell OpenRD (Base|Client) Board Setup
+ * Marvell OpenRD (Base|Client|Ultimate) Board Setup
*
* This file is licensed under the terms of the GNU General Public
* License version 2. This program is licensed "as is" without any
@@ -73,9 +73,15 @@ static void __init openrd_init(void)

kirkwood_ehci_init();

+ if (machine_is_openrd_ultimate()) {
+ openrd_ge00_data.phy_addr = MV643XX_ETH_PHY_ADDR(0);
+ openrd_ge01_data.phy_addr = MV643XX_ETH_PHY_ADDR(1);
+ }
+
kirkwood_ge00_init(&openrd_ge00_data);
- if (machine_is_openrd_client())
+ if (!machine_is_openrd_base())
kirkwood_ge01_init(&openrd_ge01_data);
+
kirkwood_sata_init(&openrd_sata_data);
kirkwood_sdio_init(&openrd_mvsdio_data);

@@ -84,8 +90,7 @@ static void __init openrd_init(void)

static int __init openrd_pci_init(void)
{
- if (machine_is_openrd_base() || machine_is_openrd_client())
- kirkwood_pcie_init();
+ kirkwood_pcie_init();

return 0;
}
@@ -116,3 +121,16 @@ MACHINE_START(OPENRD_CLIENT, "Marvell OpenRD Client Board")
.timer = &kirkwood_timer,
MACHINE_END
#endif
+
+#ifdef CONFIG_MACH_OPENRD_ULTIMATE
+MACHINE_START(OPENRD_ULTIMATE, "Marvell OpenRD Ultimate Board")
+ /* Maintainer: Dhaval Vasa <[email protected]> */
+ .phys_io = KIRKWOOD_REGS_PHYS_BASE,
+ .io_pg_offst = ((KIRKWOOD_REGS_VIRT_BASE) >> 18) & 0xfffc,
+ .boot_params = 0x00000100,
+ .init_machine = openrd_init,
+ .map_io = kirkwood_map_io,
+ .init_irq = kirkwood_init_irq,
+ .timer = &kirkwood_timer,
+MACHINE_END
+#endif
--
1.7.1


2010-06-10 05:50:29

by Alexander Clouter

[permalink] [raw]
Subject: Re: [PATCH] Marvell OpenRD-Ultimate machine support

Hi,

* Dmytro Milinevskyy <[email protected]> [2010-06-10 05:24:56+0300]:
>
> [snipped]
>
> @@ -84,8 +90,7 @@ static void __init openrd_init(void)
>
> static int __init openrd_pci_init(void)
> {
> - if (machine_is_openrd_base() || machine_is_openrd_client())
> - kirkwood_pcie_init();
> + kirkwood_pcie_init();
>
> return 0;
> }
>
You just 'broke' kirkwood, welcome to the club. We all have done this:

http://marc.info/?l=linux-arm-kernel&m=125624586625065&w=2
http://marc.info/?l=linux-arm-kernel&m=125804222219159&w=2

Put that 'if()' clause back in, the 'subsys_initcall(openrd_pci_init)'
causes that code to run on *all* compiled in boards (think of the case
where you have SheevaPlug's also enabled).

Cheers

--
Alexander Clouter
.sigmonster says: Poverty begins at home.

2010-06-10 06:15:47

by Dmytro Milinevskyy

[permalink] [raw]
Subject: Re: [PATCH] Marvell OpenRD-Ultimate machine support

Hi!

You are right, I didn't notice that it's called via subsys_initcall.

-- Dima

On Thu, Jun 10, 2010 at 8:50 AM, Alexander Clouter <[email protected]> wrote:
> Hi,
>
> * Dmytro Milinevskyy <[email protected]> [2010-06-10 05:24:56+0300]:
>>
>> [snipped]
>>
>> @@ -84,8 +90,7 @@ static void __init openrd_init(void)
>>
>> ?static int __init openrd_pci_init(void)
>> ?{
>> - ? ? if (machine_is_openrd_base() || machine_is_openrd_client())
>> - ? ? ? ? ? ? kirkwood_pcie_init();
>> + ? ? kirkwood_pcie_init();
>>
>> ? ? ? return 0;
>> ?}
>>
> You just 'broke' kirkwood, welcome to the club. ?We all have done this:
>
> http://marc.info/?l=linux-arm-kernel&m=125624586625065&w=2
> http://marc.info/?l=linux-arm-kernel&m=125804222219159&w=2
>
> Put that 'if()' clause back in, the 'subsys_initcall(openrd_pci_init)'
> causes that code to run on *all* compiled in boards (think of the case
> where you have SheevaPlug's also enabled).
>
> Cheers
>
> --
> Alexander Clouter
> .sigmonster says: Poverty begins at home.
>