Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[PATCH v7 2/5] ARM: tegra: add support for Trusted Foundations

1 view
Skip to first unread message

Alexandre Courbot

unread,
Oct 4, 2013, 12:40:01 PM10/4/13
to
Register the firmware operations for Trusted Foundations if the device
tree indicates it is active on the device.

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
Documentation/devicetree/bindings/arm/tegra.txt | 5 +++++
arch/arm/mach-tegra/Kconfig | 1 +
arch/arm/mach-tegra/common.c | 2 ++
3 files changed, 8 insertions(+)

diff --git a/Documentation/devicetree/bindings/arm/tegra.txt b/Documentation/devicetree/bindings/arm/tegra.txt
index ed9c853..8e51884 100644
--- a/Documentation/devicetree/bindings/arm/tegra.txt
+++ b/Documentation/devicetree/bindings/arm/tegra.txt
@@ -32,3 +32,8 @@ board-specific compatible values:
nvidia,whistler
toradex,colibri_t20-512
toradex,iris
+
+Trusted Foundations
+-------------------------------------------
+Tegra supports the Trusted Foundation secure monitor. See the
+"tl,trusted-foundations" binding's documentation for more details.
diff --git a/arch/arm/mach-tegra/Kconfig b/arch/arm/mach-tegra/Kconfig
index 67a76f2..4e51278 100644
--- a/arch/arm/mach-tegra/Kconfig
+++ b/arch/arm/mach-tegra/Kconfig
@@ -2,6 +2,7 @@ config ARCH_TEGRA
bool "NVIDIA Tegra" if ARCH_MULTI_V7
select ARCH_HAS_CPUFREQ
select ARCH_REQUIRE_GPIOLIB
+ select ARCH_SUPPORTS_TRUSTED_FOUNDATIONS
select ARM_GIC
select CLKDEV_LOOKUP
select CLKSRC_MMIO
diff --git a/arch/arm/mach-tegra/common.c b/arch/arm/mach-tegra/common.c
index 94a119a..b405e45 100644
--- a/arch/arm/mach-tegra/common.c
+++ b/arch/arm/mach-tegra/common.c
@@ -27,6 +27,7 @@
#include <linux/clk-provider.h>

#include <asm/hardware/cache-l2x0.h>
+#include <asm/trusted_foundations.h>

#include "board.h"
#include "common.h"
@@ -99,6 +100,7 @@ static void __init tegra_init_cache(void)

void __init tegra_init_early(void)
{
+ of_register_trusted_foundations();
tegra_cpu_reset_handler_init();
tegra_apb_io_init();
tegra_init_fuse();
--
1.8.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Alexandre Courbot

unread,
Oct 4, 2013, 12:40:02 PM10/4/13
to
Hopefully final version of this patchset. If I'm not mistaken the last thing
that prevents Stephen from merging it is Russel's Ack. Russel, could you check
it?

Trusted Foundations is an ARM secure monitor that is used by many Tegra-based
retail devices (like SHIELD). Adding support to it allows these devices to boot
on a purely upstream kernel.

Public documentation about Trusted Foundations is scarce, but over the various
versions of this patch it has been established that it does not follow the ARM
SMC convention and also has no connection whatsoever with PSCI. As such, it
requires an implementation of its own and the firmware interface introduced by
Tomasz provides an adequate framework for it.

Changes since v6:
- Improved and simplified header conditionals when TF support is compiled in
or not as advised by Stephen.

Alexandre Courbot (5):
ARM: add basic support for Trusted Foundations
ARM: tegra: add support for Trusted Foundations
ARM: tegra: split setting of CPU reset handler
ARM: tegra: set CPU reset handler with firmware op
ARM: tegra: support Trusted Foundations by default

.../arm/firmware/tl,trusted-foundations.txt | 17 +++++
Documentation/devicetree/bindings/arm/tegra.txt | 5 ++
.../devicetree/bindings/vendor-prefixes.txt | 1 +
arch/arm/Kconfig | 2 +
arch/arm/Makefile | 1 +
arch/arm/configs/tegra_defconfig | 1 +
arch/arm/firmware/Kconfig | 28 ++++++++
arch/arm/firmware/Makefile | 1 +
arch/arm/firmware/trusted_foundations.c | 77 ++++++++++++++++++++++
arch/arm/include/asm/trusted_foundations.h | 64 ++++++++++++++++++
arch/arm/mach-tegra/Kconfig | 1 +
arch/arm/mach-tegra/common.c | 2 +
arch/arm/mach-tegra/reset.c | 40 +++++++----
13 files changed, 229 insertions(+), 11 deletions(-)
create mode 100644 Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt
create mode 100644 arch/arm/firmware/Kconfig
create mode 100644 arch/arm/firmware/Makefile
create mode 100644 arch/arm/firmware/trusted_foundations.c
create mode 100644 arch/arm/include/asm/trusted_foundations.h

Alexandre Courbot

unread,
Oct 4, 2013, 12:40:02 PM10/4/13
to
Trusted Foundations is a TrustZone-based secure monitor for ARM that
can be invoked using the same SMC-based API on all supported
platforms. This patch adds initial basic support for Trusted
Foundations using the ARM firmware API. Current features are limited
to the ability to boot secondary processors.

Note: The API followed by Trusted Foundations does *not* follow the SMC
calling conventions. It has nothing to do with PSCI neither and is only
relevant to devices that use Trusted Foundations (like most Tegra-based
retail devices).

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
.../arm/firmware/tl,trusted-foundations.txt | 17 +++++
.../devicetree/bindings/vendor-prefixes.txt | 1 +
arch/arm/Kconfig | 2 +
arch/arm/Makefile | 1 +
arch/arm/firmware/Kconfig | 28 ++++++++
arch/arm/firmware/Makefile | 1 +
arch/arm/firmware/trusted_foundations.c | 77 ++++++++++++++++++++++
arch/arm/include/asm/trusted_foundations.h | 64 ++++++++++++++++++
8 files changed, 191 insertions(+)
create mode 100644 Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt
create mode 100644 arch/arm/firmware/Kconfig
create mode 100644 arch/arm/firmware/Makefile
create mode 100644 arch/arm/firmware/trusted_foundations.c
create mode 100644 arch/arm/include/asm/trusted_foundations.h

diff --git a/Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt b/Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt
new file mode 100644
index 0000000..3954bbd
--- /dev/null
+++ b/Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt
@@ -0,0 +1,17 @@
+Trusted Foundations
+
+Boards that use the Trusted Foundations secure monitor can signal its
+presence by declaring a node compatible with "tl,trusted-foundations"
+under the root node.
+
+Required properties:
+- compatible : "tl,trusted-foundations"
+- version-major : major version number of Trusted Foundations firmware
+- version-minor: minor version number of Trusted Foundations firmware
+
+Example:
+ firmware {
+ compatible = "tl,trusted-foundations";
+ version-major = <2>;
+ version-minor = <8>;
+ };
diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt
index 2956800..a374eaa 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.txt
+++ b/Documentation/devicetree/bindings/vendor-prefixes.txt
@@ -66,6 +66,7 @@ ste ST-Ericsson
stericsson ST-Ericsson
toumaz Toumaz
ti Texas Instruments
+tl Trusted Logic
toshiba Toshiba Corporation
v3 V3 Semiconductor
via VIA Technologies, Inc.
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 1ad6fb6..bf14cec 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1076,6 +1076,8 @@ config ARM_TIMER_SP804
select CLKSRC_MMIO
select CLKSRC_OF if OF

+source "arch/arm/firmware/Kconfig"
+
source arch/arm/mm/Kconfig

config ARM_NR_BANKS
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index a37a50f..f26712c 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -268,6 +268,7 @@ core-$(CONFIG_KVM_ARM_HOST) += arch/arm/kvm/
core-y += arch/arm/kernel/ arch/arm/mm/ arch/arm/common/
core-y += arch/arm/net/
core-y += arch/arm/crypto/
+core-y += arch/arm/firmware/
core-y += $(machdirs) $(platdirs)

drivers-$(CONFIG_OPROFILE) += arch/arm/oprofile/
diff --git a/arch/arm/firmware/Kconfig b/arch/arm/firmware/Kconfig
new file mode 100644
index 0000000..bb00ccf
--- /dev/null
+++ b/arch/arm/firmware/Kconfig
@@ -0,0 +1,28 @@
+config ARCH_SUPPORTS_FIRMWARE
+ bool
+
+config ARCH_SUPPORTS_TRUSTED_FOUNDATIONS
+ bool
+ select ARCH_SUPPORTS_FIRMWARE
+
+menu "Firmware options"
+ depends on ARCH_SUPPORTS_FIRMWARE
+
+config TRUSTED_FOUNDATIONS
+ bool "Trusted Foundations secure monitor support"
+ depends on ARCH_SUPPORTS_TRUSTED_FOUNDATIONS
+ help
+ Some devices (including most Tegra-based consumer devices on the
+ market) are booted with the Trusted Foundations secure monitor
+ active, requiring some core operations to be performed by the secure
+ monitor instead of the kernel.
+
+ This option allows the kernel to invoke the secure monitor whenever
+ required on devices using Trusted Foundations. See
+ arch/arm/include/asm/trusted_foundations.h or the
+ tl,trusted-foundations device tree binding documentation for details
+ on how to use it.
+
+ Say n if you don't know what this is about.
+
+endmenu
diff --git a/arch/arm/firmware/Makefile b/arch/arm/firmware/Makefile
new file mode 100644
index 0000000..a71f165
--- /dev/null
+++ b/arch/arm/firmware/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_TRUSTED_FOUNDATIONS) += trusted_foundations.o
diff --git a/arch/arm/firmware/trusted_foundations.c b/arch/arm/firmware/trusted_foundations.c
new file mode 100644
index 0000000..c93c5e0
--- /dev/null
+++ b/arch/arm/firmware/trusted_foundations.c
@@ -0,0 +1,77 @@
+/*
+ * Trusted Foundations support for ARM CPUs
+ *
+ * Copyright (c) 2013, NVIDIA Corporation.
+ *
+ * This program 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 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/kernel.h>
+#include <linux/init.h>
+#include <linux/of.h>
+#include <asm/firmware.h>
+#include <asm/trusted_foundations.h>
+
+#define TF_SET_CPU_BOOT_ADDR_SMC 0xfffff200
+
+static void __naked tf_generic_smc(u32 type, u32 subtype, u32 arg)
+{
+ asm volatile(
+ ".arch_extension sec\n\t"
+ "stmfd sp!, {r4 - r11, lr}\n\t"
+ __asmeq("%0", "r0")
+ __asmeq("%1", "r1")
+ __asmeq("%2", "r2")
+ "mov r3, #0\n\t"
+ "mov r4, #0\n\t"
+ "smc #0\n\t"
+ "ldmfd sp!, {r4 - r11, pc}"
+ :
+ : "r" (type), "r" (subtype), "r" (arg)
+ : "memory");
+}
+
+static int tf_set_cpu_boot_addr(int cpu, unsigned long boot_addr)
+{
+ tf_generic_smc(TF_SET_CPU_BOOT_ADDR_SMC, boot_addr, 0);
+
+ return 0;
+}
+
+static const struct firmware_ops trusted_foundations_ops = {
+ .set_cpu_boot_addr = tf_set_cpu_boot_addr,
+};
+
+void register_trusted_foundations(struct trusted_foundations_platform_data *pd)
+{
+ /* we are not using version information for now since currently
+ * supported SMCs are compatible with all TF releases */
+ register_firmware_ops(&trusted_foundations_ops);
+}
+
+void of_register_trusted_foundations(void)
+{
+ struct device_node *node;
+ struct trusted_foundations_platform_data pdata;
+ int err;
+
+ node = of_find_compatible_node(NULL, NULL, "tl,trusted-foundations");
+ if (!node)
+ return;
+
+ err = of_property_read_u32(node, "version-major", &pdata.version_major);
+ if (err != 0)
+ panic("Trusted Foundation: missing version-major property\n");
+ err = of_property_read_u32(node, "version-minor", &pdata.version_minor);
+ if (err != 0)
+ panic("Trusted Foundation: missing version-minor property\n");
+ register_trusted_foundations(&pdata);
+}
diff --git a/arch/arm/include/asm/trusted_foundations.h b/arch/arm/include/asm/trusted_foundations.h
new file mode 100644
index 0000000..749edbf
--- /dev/null
+++ b/arch/arm/include/asm/trusted_foundations.h
@@ -0,0 +1,64 @@
+/*
+ * Copyright (c) 2013, NVIDIA Corporation.
+ *
+ * This program 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 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.
+ */
+
+/*
+ * Support for the Trusted Foundations secure monitor.
+ *
+ * Trusted Foundation comes active on some ARM consumer devices (most
+ * Tegra-based devices sold on the market are concerned). Such devices can only
+ * perform some basic operations, like setting the CPU reset vector, through
+ * SMC calls to the secure monitor. The calls are completely specific to
+ * Trusted Foundations, and do *not* follow the SMC calling convention or the
+ * PSCI standard.
+ */
+
+#ifndef __ASM_ARM_TRUSTED_FOUNDATIONS_H
+#define __ASM_ARM_TRUSTED_FOUNDATIONS_H
+
+#include <linux/kconfig.h>
+
+struct trusted_foundations_platform_data {
+ unsigned int version_major;
+ unsigned int version_minor;
+};
+
+#if IS_ENABLED(CONFIG_TRUSTED_FOUNDATIONS)
+
+void register_trusted_foundations(struct trusted_foundations_platform_data *pd);
+void of_register_trusted_foundations(void);
+
+#else /* CONFIG_TRUSTED_FOUNDATIONS */
+
+#include <linux/printk.h>
+#include <linux/of.h>
+#include <asm/bug.h>
+
+static inline void register_trusted_foundations(
+ struct trusted_foundations_platform_data *pd)
+{
+ /* If we try to register TF, this means the system needs it to continue.
+ * Its absence if thus a fatal error. */
+ panic("No support for Trusted Foundations, stopping...\n");
+}
+
+static inline void of_register_trusted_foundations(void)
+{
+ /* If we find the target should enable TF but does not support it,
+ * fail as the system won't be able to do much anyway */
+ if (of_find_compatible_node(NULL, NULL, "tl,trusted-foundations"))
+ register_trusted_foundations(NULL);
+}
+#endif /* CONFIG_TRUSTED_FOUNDATIONS */
+
+#endif

Alexandre Courbot

unread,
Oct 4, 2013, 12:40:02 PM10/4/13
to
Not all Tegra devices can set the CPU reset handler in the same way.
In particular, devices using a TrustZone secure monitor cannot set the
reset handler directly and need to do it through a firmware operation.

This patch separates the act of setting the reset handler from its
preparation, so the former can be implemented in a different way.

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
arch/arm/mach-tegra/reset.c | 27 +++++++++++++++++----------
1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/arch/arm/mach-tegra/reset.c b/arch/arm/mach-tegra/reset.c
index fd0bbf8..e282395 100644
--- a/arch/arm/mach-tegra/reset.c
+++ b/arch/arm/mach-tegra/reset.c
@@ -33,26 +33,18 @@

static bool is_enabled;

-static void __init tegra_cpu_reset_handler_enable(void)
+static void __init tegra_cpu_reset_handler_set(const u32 reset_address)
{
- void __iomem *iram_base = IO_ADDRESS(TEGRA_IRAM_RESET_BASE);
void __iomem *evp_cpu_reset =
IO_ADDRESS(TEGRA_EXCEPTION_VECTORS_BASE + 0x100);
void __iomem *sb_ctrl = IO_ADDRESS(TEGRA_SB_BASE);
u32 reg;

- BUG_ON(is_enabled);
- BUG_ON(tegra_cpu_reset_handler_size > TEGRA_IRAM_RESET_HANDLER_SIZE);
-
- memcpy(iram_base, (void *)__tegra_cpu_reset_handler_start,
- tegra_cpu_reset_handler_size);
-
/*
* NOTE: This must be the one and only write to the EVP CPU reset
* vector in the entire system.
*/
- writel(TEGRA_IRAM_RESET_BASE + tegra_cpu_reset_handler_offset,
- evp_cpu_reset);
+ writel(reset_address, evp_cpu_reset);
wmb();
reg = readl(evp_cpu_reset);

@@ -66,6 +58,21 @@ static void __init tegra_cpu_reset_handler_enable(void)
writel(reg, sb_ctrl);
wmb();
}
+}
+
+static void __init tegra_cpu_reset_handler_enable(void)
+{
+ void __iomem *iram_base = IO_ADDRESS(TEGRA_IRAM_RESET_BASE);
+ const u32 reset_address = TEGRA_IRAM_RESET_BASE +
+ tegra_cpu_reset_handler_offset;
+
+ BUG_ON(is_enabled);
+ BUG_ON(tegra_cpu_reset_handler_size > TEGRA_IRAM_RESET_HANDLER_SIZE);
+
+ memcpy(iram_base, (void *)__tegra_cpu_reset_handler_start,
+ tegra_cpu_reset_handler_size);
+
+ tegra_cpu_reset_handler_set(reset_address);

is_enabled = true;

Alexandre Courbot

unread,
Oct 4, 2013, 12:40:03 PM10/4/13
to
Support for Trusted Foundations is light and allows the kernel to run on
a wider range of devices, so enable it by default.

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
arch/arm/configs/tegra_defconfig | 1 +
1 file changed, 1 insertion(+)

diff --git a/arch/arm/configs/tegra_defconfig b/arch/arm/configs/tegra_defconfig
index ea042e8..dddb97b 100644
--- a/arch/arm/configs/tegra_defconfig
+++ b/arch/arm/configs/tegra_defconfig
@@ -32,6 +32,7 @@ CONFIG_PCI=y
CONFIG_PCI_MSI=y
CONFIG_PCI_TEGRA=y
CONFIG_PCIEPORTBUS=y
+CONFIG_TRUSTED_FOUNDATIONS=y
CONFIG_SMP=y
CONFIG_PREEMPT=y
CONFIG_AEABI=y

Alexandre Courbot

unread,
Oct 4, 2013, 12:40:02 PM10/4/13
to
Use a firmware operation to set the CPU reset handler and only resort to
doing it ourselves if there is none defined.

This supports the booting of secondary CPUs on devices using a TrustZone
secure monitor.

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
arch/arm/mach-tegra/reset.c | 17 ++++++++++++++---
1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/arch/arm/mach-tegra/reset.c b/arch/arm/mach-tegra/reset.c
index e282395..368af0a 100644
--- a/arch/arm/mach-tegra/reset.c
+++ b/arch/arm/mach-tegra/reset.c
@@ -21,6 +21,7 @@

#include <asm/cacheflush.h>
#include <asm/hardware/cache-l2x0.h>
+#include <asm/firmware.h>

#include "iomap.h"
#include "irammap.h"
@@ -65,6 +66,7 @@ static void __init tegra_cpu_reset_handler_enable(void)
void __iomem *iram_base = IO_ADDRESS(TEGRA_IRAM_RESET_BASE);
const u32 reset_address = TEGRA_IRAM_RESET_BASE +
tegra_cpu_reset_handler_offset;
+ int err;

BUG_ON(is_enabled);
BUG_ON(tegra_cpu_reset_handler_size > TEGRA_IRAM_RESET_HANDLER_SIZE);
@@ -72,9 +74,18 @@ static void __init tegra_cpu_reset_handler_enable(void)
memcpy(iram_base, (void *)__tegra_cpu_reset_handler_start,
tegra_cpu_reset_handler_size);

- tegra_cpu_reset_handler_set(reset_address);
-
- is_enabled = true;
+ err = call_firmware_op(set_cpu_boot_addr, 0, reset_address);
+ switch (err) {
+ case -ENOSYS:
+ tegra_cpu_reset_handler_set(reset_address);
+ /* pass-through */
+ case 0:
+ is_enabled = true;
+ break;
+ default:
+ pr_crit("Cannot set CPU reset handler: %d\n", err);
+ BUG();
+ }
}

void __init tegra_cpu_reset_handler_init(void)

Alex Courbot

unread,
Oct 6, 2013, 2:20:03 PM10/6/13
to
On 10/04/2013 09:37 AM, Alexandre Courbot wrote:
> Hopefully final version of this patchset. If I'm not mistaken the last thing
> that prevents Stephen from merging it is Russel's Ack. Russel, could you check
> it?

Oops.

s/Russel/Russell/g, with apologies for the continuous misspelling. m(__)m

Alex.

Kevin Hilman

unread,
Oct 7, 2013, 2:20:01 PM10/7/13
to
Alexandre Courbot <acou...@nvidia.com> writes:

> Trusted Foundations is a TrustZone-based secure monitor for ARM that
> can be invoked using the same SMC-based API on all supported
> platforms. This patch adds initial basic support for Trusted
> Foundations using the ARM firmware API. Current features are limited
> to the ability to boot secondary processors.
>
> Note: The API followed by Trusted Foundations does *not* follow the SMC
> calling conventions. It has nothing to do with PSCI neither and is only
> relevant to devices that use Trusted Foundations (like most Tegra-based
> retail devices).
>
> Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
> Reviewed-by: Tomasz Figa <t.f...@samsung.com>
> Reviewed-by: Stephen Warren <swa...@nvidia.com>

[...]

> +void register_trusted_foundations(struct trusted_foundations_platform_data *pd)
> +{
> + /* we are not using version information for now since currently
> + * supported SMCs are compatible with all TF releases */

nit: multi-line comment style...

[...]

> +static inline void register_trusted_foundations(
> + struct trusted_foundations_platform_data *pd)
> +{
> + /* If we try to register TF, this means the system needs it to continue.
> + * Its absence if thus a fatal error. */

...and here...

> + panic("No support for Trusted Foundations, stopping...\n");
> +}
> +
> +static inline void of_register_trusted_foundations(void)
> +{
> + /* If we find the target should enable TF but does not support it,
> + * fail as the system won't be able to do much anyway */

...and here.

> + if (of_find_compatible_node(NULL, NULL, "tl,trusted-foundations"))
> + register_trusted_foundations(NULL);
> +}
> +#endif /* CONFIG_TRUSTED_FOUNDATIONS */
> +
> +#endif

Kevin

Stephen Warren

unread,
Oct 7, 2013, 3:30:03 PM10/7/13
to
On 10/04/2013 10:37 AM, Alexandre Courbot wrote:
> Trusted Foundations is a TrustZone-based secure monitor for ARM that
> can be invoked using the same SMC-based API on all supported
> platforms. This patch adds initial basic support for Trusted
> Foundations using the ARM firmware API. Current features are limited
> to the ability to boot secondary processors.
>
> Note: The API followed by Trusted Foundations does *not* follow the SMC
> calling conventions. It has nothing to do with PSCI neither and is only
> relevant to devices that use Trusted Foundations (like most Tegra-based
> retail devices).

Hmmm. This patch didn't get sent to the DT maintainers. I'm CC'ing them
all in this reply, and hence quoting the two relevant files in full. I'm
looking for an ack to take this patch through the Tegra tree.

I'm also looking for an ack from perhaps Russell on the code changes in
the patch.

Michal Simek

unread,
Oct 8, 2013, 4:20:02 AM10/8/13
to
Hi,
This is just another example how to add sw description to dts.
I have briefly looked at tegra20.dtsi and there are IPs like timer, intc, etc
which are in the DTS without any bus.

Add this firmware node to the root is just +1 case to this mess
but IMHO will be good to have specific DT part which will be used for
this sw setting which are coming to DTS all the time.

Grant, Rob: Where is the proper location for these type of description?

Thanks,
Michal




--
Michal Simek, Ing. (M.Eng), OpenPGP -> KeyID: FE3D1F91
w: www.monstr.eu p: +42-0-721842854
Maintainer of Linux kernel - Microblaze cpu - http://www.monstr.eu/fdt/
Maintainer of Linux kernel - Xilinx Zynq ARM architecture
Microblaze U-BOOT custodian and responsible for u-boot arm zynq platform


signature.asc

Olof Johansson

unread,
Oct 9, 2013, 7:50:02 PM10/9/13
to
Usually we've been using the /firmware hierarchy for these kind of things.

Chrome OS uses /firmware/chromeos for some of the information passing
from FW to the OS.

You might want a separate node under /firmware for this, since this is
"just" the secure runtime portion of things, there might be need to
add things for/from the bootloader too.


-Olof

Olof Johansson

unread,
Oct 9, 2013, 7:50:02 PM10/9/13
to
The latter function can just be shortcut unto a local panic too.

But do we really need to panic here? It's a PITA to debug since it'll
be a very early panic. I'd rather have the system log errors in dmesg
but continue booting with one cpu.

(That can be done incrementally on top of this, and doesn't need to
hold up this patch series).


-Olof

Michal Simek

unread,
Oct 10, 2013, 1:20:02 AM10/10/13
to
Is this documented somewhere? I mean that firmware node should be used for this.
signature.asc

Alexandre Courbot

unread,
Oct 10, 2013, 5:00:02 PM10/10/13
to
Actually the first version of this series continued on one CPU if TF
support was missing (as secondary CPUs would simply fail to boot), but
IIRC Stephen advocated for the current behavior (panic) instead. I
agree such an early panic is hard to debug, especially considering
that most devices using TF (Tegra retail devices) do not even feature
a serial port.

On the other hand, if we continue booting in these conditions the
system freezes as soon as cpuidle kicks in, so we won't be going very
far anyway. However the console is up when it happens.

Maybe if there is a way to disable the kernel features TF is supposed
to manage (SMP & cpuidle at the moment) when registration fails, the
system could continue more gracefully (even if on one leg) and warn
the user about this condition. Stephen, can I have your thoughts on
this?

Kevin Hilman

unread,
Oct 10, 2013, 5:20:02 PM10/10/13
to
Alexandre Courbot <gnu...@gmail.com> writes:

> On Wed, Oct 9, 2013 at 4:47 PM, Olof Johansson <ol...@lixom.net> wrote:
>> On Fri, Oct 4, 2013 at 9:37 AM, Alexandre Courbot <acou...@nvidia.com> wrote:

[...]
cpu_idle_poll_ctrl(true) should solve that problem.

Kevin

Alexandre Courbot

unread,
Oct 10, 2013, 6:30:01 PM10/10/13
to
On Thu, Oct 10, 2013 at 3:22 PM, Stephen Warren <swa...@wwwdotorg.org> wrote:
> On 10/10/2013 02:58 PM, Alexandre Courbot wrote:
>> On Wed, Oct 9, 2013 at 4:47 PM, Olof Johansson <ol...@lixom.net> wrote:
>>> On Fri, Oct 4, 2013 at 9:37 AM, Alexandre Courbot <acou...@nvidia.com> wrote:
> ...
>>> But do we really need to panic here? It's a PITA to debug since it'll
>>> be a very early panic. I'd rather have the system log errors in dmesg
>>> but continue booting with one cpu.
>>
>> Actually the first version of this series continued on one CPU if TF
>> support was missing (as secondary CPUs would simply fail to boot), but
>> IIRC Stephen advocated for the current behavior (panic) instead. I
>> agree such an early panic is hard to debug, especially considering
>> that most devices using TF (Tegra retail devices) do not even feature
>> a serial port.
>>
>> On the other hand, if we continue booting in these conditions the
>> system freezes as soon as cpuidle kicks in, so we won't be going very
>> far anyway. However the console is up when it happens.
>>
>> Maybe if there is a way to disable the kernel features TF is supposed
>> to manage (SMP & cpuidle at the moment) when registration fails, the
>> system could continue more gracefully (even if on one leg) and warn
>> the user about this condition. Stephen, can I have your thoughts on
>> this?
>
> If we can disable the features that rely on TF/secure-mode when support
> is missing, then doing that rather than panic is fine. We do something
> similar for PCIe on Tegra20, where LP2 cpuidle state must be disabled or
> PCIe interrupts won't work.
>
> I would be inclined to merge the patch as-is right now, since it's only
> enabling a new feature that won't see any regressions due to this panic,
> and then add the disable-other-features with a separate patch/series
> later. Does that seem reasonable?

According to Olof we should also be moving the DT node under
/firmware, so I guess another revision is in order to at least get
this part right.

3.12 also added a powered-down cpuidle state for Tegra which freezes
the system if TF is enabled as the firmware driver does not support
idle yet, but this can be fixed later.

Olof Johansson

unread,
Oct 10, 2013, 6:30:02 PM10/10/13
to
On Thu, Oct 10, 2013 at 3:22 PM, Stephen Warren <swa...@wwwdotorg.org> wrote:
> On 10/10/2013 02:58 PM, Alexandre Courbot wrote:
>> On Wed, Oct 9, 2013 at 4:47 PM, Olof Johansson <ol...@lixom.net> wrote:
>>> On Fri, Oct 4, 2013 at 9:37 AM, Alexandre Courbot <acou...@nvidia.com> wrote:
> ...
>>> But do we really need to panic here? It's a PITA to debug since it'll
>>> be a very early panic. I'd rather have the system log errors in dmesg
>>> but continue booting with one cpu.
>>
>> Actually the first version of this series continued on one CPU if TF
>> support was missing (as secondary CPUs would simply fail to boot), but
>> IIRC Stephen advocated for the current behavior (panic) instead. I
>> agree such an early panic is hard to debug, especially considering
>> that most devices using TF (Tegra retail devices) do not even feature
>> a serial port.
>>
>> On the other hand, if we continue booting in these conditions the
>> system freezes as soon as cpuidle kicks in, so we won't be going very
>> far anyway. However the console is up when it happens.
>>
>> Maybe if there is a way to disable the kernel features TF is supposed
>> to manage (SMP & cpuidle at the moment) when registration fails, the
>> system could continue more gracefully (even if on one leg) and warn
>> the user about this condition. Stephen, can I have your thoughts on
>> this?
>
> If we can disable the features that rely on TF/secure-mode when support
> is missing, then doing that rather than panic is fine. We do something
> similar for PCIe on Tegra20, where LP2 cpuidle state must be disabled or
> PCIe interrupts won't work.
>
> I would be inclined to merge the patch as-is right now, since it's only
> enabling a new feature that won't see any regressions due to this panic,
> and then add the disable-other-features with a separate patch/series
> later. Does that seem reasonable?

Yes, I tried to make that clear with my initial feedback: I think it
would be good to get away from the panic, but we can merge with it now
and fix it with later patches.


-Olof

Stephen Warren

unread,
Oct 10, 2013, 6:30:02 PM10/10/13
to
On 10/10/2013 02:58 PM, Alexandre Courbot wrote:
> On Wed, Oct 9, 2013 at 4:47 PM, Olof Johansson <ol...@lixom.net> wrote:
>> On Fri, Oct 4, 2013 at 9:37 AM, Alexandre Courbot <acou...@nvidia.com> wrote:
...
>> But do we really need to panic here? It's a PITA to debug since it'll
>> be a very early panic. I'd rather have the system log errors in dmesg
>> but continue booting with one cpu.
>
> Actually the first version of this series continued on one CPU if TF
> support was missing (as secondary CPUs would simply fail to boot), but
> IIRC Stephen advocated for the current behavior (panic) instead. I
> agree such an early panic is hard to debug, especially considering
> that most devices using TF (Tegra retail devices) do not even feature
> a serial port.
>
> On the other hand, if we continue booting in these conditions the
> system freezes as soon as cpuidle kicks in, so we won't be going very
> far anyway. However the console is up when it happens.
>
> Maybe if there is a way to disable the kernel features TF is supposed
> to manage (SMP & cpuidle at the moment) when registration fails, the
> system could continue more gracefully (even if on one leg) and warn
> the user about this condition. Stephen, can I have your thoughts on
> this?

If we can disable the features that rely on TF/secure-mode when support
is missing, then doing that rather than panic is fine. We do something
similar for PCIe on Tegra20, where LP2 cpuidle state must be disabled or
PCIe interrupts won't work.

I would be inclined to merge the patch as-is right now, since it's only
enabling a new feature that won't see any regressions due to this panic,
and then add the disable-other-features with a separate patch/series
later. Does that seem reasonable?

Alexandre Courbot

unread,
Oct 11, 2013, 5:50:02 PM10/11/13
to
Use a firmware operation to set the CPU reset handler and only resort to
doing it ourselves if there is none defined.

This supports the booting of secondary CPUs on devices using a TrustZone
secure monitor.

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---

Alexandre Courbot

unread,
Oct 11, 2013, 5:50:02 PM10/11/13
to
Hopefully this has received enough review to be merged (although a few more
acks/review especially from DT people on patch 1 would be nice). Thanks
everyone for all the input and comments!

Trusted Foundations is an ARM secure monitor that is used by many Tegra-based
retail devices (like SHIELD and TegraNote). Adding support to it allows these
devices to boot on a purely upstream kernel.

Public documentation about Trusted Foundations is scarce, but over the various
versions of this patch it has been established that it does not follow the ARM
SMC convention and also has no connection whatsoever with PSCI. As such, it
requires an implementation of its own and the firmware interface introduced by
Tomasz provides an adequate framework for it.

Changes since v7:
- explicitly request the TF DT node to be under /firmware/ in the bindings
documentation
- renamed arguments of tf_generic_smc() (suggestion made by Linus W. some time
ago that I overlooked).
- fixed multi-line comment style

TODO in a future patchset:
- disable TF-dependant features (SMP, cpuidle) if TF cannot be supported instead
or doing an early panic
- complete TF support with idle and L2 cache handling (ongoing)

Alexandre Courbot (5):
ARM: add basic support for Trusted Foundations
ARM: tegra: add support for Trusted Foundations
ARM: tegra: split setting of CPU reset handler
ARM: tegra: set CPU reset handler with firmware op
ARM: tegra: support Trusted Foundations by default

.../arm/firmware/tl,trusted-foundations.txt | 20 ++++++
Documentation/devicetree/bindings/arm/tegra.txt | 5 ++
.../devicetree/bindings/vendor-prefixes.txt | 1 +
arch/arm/Kconfig | 2 +
arch/arm/Makefile | 1 +
arch/arm/configs/tegra_defconfig | 1 +
arch/arm/firmware/Kconfig | 28 ++++++++
arch/arm/firmware/Makefile | 1 +
arch/arm/firmware/trusted_foundations.c | 79 ++++++++++++++++++++++
arch/arm/include/asm/trusted_foundations.h | 68 +++++++++++++++++++
arch/arm/mach-tegra/Kconfig | 1 +
arch/arm/mach-tegra/common.c | 2 +
arch/arm/mach-tegra/reset.c | 40 ++++++++---
13 files changed, 238 insertions(+), 11 deletions(-)
create mode 100644 Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt
create mode 100644 arch/arm/firmware/Kconfig
create mode 100644 arch/arm/firmware/Makefile
create mode 100644 arch/arm/firmware/trusted_foundations.c
create mode 100644 arch/arm/include/asm/trusted_foundations.h

Alexandre Courbot

unread,
Oct 11, 2013, 5:50:02 PM10/11/13
to
Not all Tegra devices can set the CPU reset handler in the same way.
In particular, devices using a TrustZone secure monitor cannot set the
reset handler directly and need to do it through a firmware operation.

This patch separates the act of setting the reset handler from its
preparation, so the former can be implemented in a different way.

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
arch/arm/mach-tegra/reset.c | 27 +++++++++++++++++----------
1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/arch/arm/mach-tegra/reset.c b/arch/arm/mach-tegra/reset.c
index fd0bbf8..e282395 100644
--- a/arch/arm/mach-tegra/reset.c
+++ b/arch/arm/mach-tegra/reset.c

Alexandre Courbot

unread,
Oct 11, 2013, 5:50:02 PM10/11/13
to
Register the firmware operations for Trusted Foundations if the device
tree indicates it is active on the device.

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
Documentation/devicetree/bindings/arm/tegra.txt | 5 +++++
arch/arm/mach-tegra/Kconfig | 1 +

Alexandre Courbot

unread,
Oct 11, 2013, 5:50:02 PM10/11/13
to
Trusted Foundations is a TrustZone-based secure monitor for ARM that
can be invoked using the same SMC-based API on all supported
platforms. This patch adds initial basic support for Trusted
Foundations using the ARM firmware API. Current features are limited
to the ability to boot secondary processors.

Note: The API followed by Trusted Foundations does *not* follow the SMC
calling conventions. It has nothing to do with PSCI neither and is only
relevant to devices that use Trusted Foundations (like most Tegra-based
retail devices).

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
.../arm/firmware/tl,trusted-foundations.txt | 20 ++++++
.../devicetree/bindings/vendor-prefixes.txt | 1 +
arch/arm/Kconfig | 2 +
arch/arm/Makefile | 1 +
arch/arm/firmware/Kconfig | 28 ++++++++
arch/arm/firmware/Makefile | 1 +
arch/arm/firmware/trusted_foundations.c | 79 ++++++++++++++++++++++
arch/arm/include/asm/trusted_foundations.h | 68 +++++++++++++++++++
8 files changed, 200 insertions(+)
create mode 100644 Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt
create mode 100644 arch/arm/firmware/Kconfig
create mode 100644 arch/arm/firmware/Makefile
create mode 100644 arch/arm/firmware/trusted_foundations.c
create mode 100644 arch/arm/include/asm/trusted_foundations.h

diff --git a/Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt b/Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt
new file mode 100644
index 0000000..2ec75c9
--- /dev/null
+++ b/Documentation/devicetree/bindings/arm/firmware/tl,trusted-foundations.txt
@@ -0,0 +1,20 @@
+Trusted Foundations
+-------------------
+
+Boards that use the Trusted Foundations secure monitor can signal its
+presence by declaring a node compatible with "tl,trusted-foundations"
+under the /firmware/ node
+
+Required properties:
+- compatible : "tl,trusted-foundations"
+- version-major : major version number of Trusted Foundations firmware
+- version-minor: minor version number of Trusted Foundations firmware
+
+Example:
+ firmware {
+ trusted-foundations {
index 0000000..fe063d5
--- /dev/null
+++ b/arch/arm/firmware/trusted_foundations.c
@@ -0,0 +1,79 @@
+/*
+ * Trusted Foundations support for ARM CPUs
+ *
+ * Copyright (c) 2013, NVIDIA Corporation.
+ *
+ * This program 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 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/kernel.h>
+#include <linux/init.h>
+#include <linux/of.h>
+#include <asm/firmware.h>
+#include <asm/trusted_foundations.h>
+
+#define TF_SET_CPU_BOOT_ADDR_SMC 0xfffff200
+
+static void __naked tf_generic_smc(u32 type, u32 arg1, u32 arg2)
+{
+ asm volatile(
+ ".arch_extension sec\n\t"
+ "stmfd sp!, {r4 - r11, lr}\n\t"
+ __asmeq("%0", "r0")
+ __asmeq("%1", "r1")
+ __asmeq("%2", "r2")
+ "mov r3, #0\n\t"
+ "mov r4, #0\n\t"
+ "smc #0\n\t"
+ "ldmfd sp!, {r4 - r11, pc}"
+ :
+ : "r" (type), "r" (arg1), "r" (arg2)
+ : "memory");
+}
+
+static int tf_set_cpu_boot_addr(int cpu, unsigned long boot_addr)
+{
+ tf_generic_smc(TF_SET_CPU_BOOT_ADDR_SMC, boot_addr, 0);
+
+ return 0;
+}
+
+static const struct firmware_ops trusted_foundations_ops = {
+ .set_cpu_boot_addr = tf_set_cpu_boot_addr,
+};
+
+void register_trusted_foundations(struct trusted_foundations_platform_data *pd)
+{
+ /*
+ * we are not using version information for now since currently
+ * supported SMCs are compatible with all TF releases
+ */
+ register_firmware_ops(&trusted_foundations_ops);
+}
+
+void of_register_trusted_foundations(void)
+{
+ struct device_node *node;
+ struct trusted_foundations_platform_data pdata;
+ int err;
+
+ node = of_find_compatible_node(NULL, NULL, "tl,trusted-foundations");
+ if (!node)
+ return;
+
+ err = of_property_read_u32(node, "version-major", &pdata.version_major);
+ if (err != 0)
+ panic("Trusted Foundation: missing version-major property\n");
+ err = of_property_read_u32(node, "version-minor", &pdata.version_minor);
+ if (err != 0)
+ panic("Trusted Foundation: missing version-minor property\n");
+ register_trusted_foundations(&pdata);
+}
diff --git a/arch/arm/include/asm/trusted_foundations.h b/arch/arm/include/asm/trusted_foundations.h
new file mode 100644
index 0000000..c6f20bd
--- /dev/null
+++ b/arch/arm/include/asm/trusted_foundations.h
@@ -0,0 +1,68 @@
+static inline void register_trusted_foundations(
+ struct trusted_foundations_platform_data *pd)
+{
+ /*
+ * If we try to register TF, this means the system needs it to continue.
+ * Its absence if thus a fatal error.
+ */
+ panic("No support for Trusted Foundations, stopping...\n");
+}
+
+static inline void of_register_trusted_foundations(void)
+{
+ /*
+ * If we find the target should enable TF but does not support it,
+ * fail as the system won't be able to do much anyway
+ */
+ if (of_find_compatible_node(NULL, NULL, "tl,trusted-foundations"))
+ register_trusted_foundations(NULL);
+}
+#endif /* CONFIG_TRUSTED_FOUNDATIONS */
+
+#endif

Alexandre Courbot

unread,
Oct 11, 2013, 5:50:03 PM10/11/13
to
Support for Trusted Foundations is light and allows the kernel to run on
a wider range of devices, so enable it by default.

Signed-off-by: Alexandre Courbot <acou...@nvidia.com>
Reviewed-by: Tomasz Figa <t.f...@samsung.com>
Reviewed-by: Stephen Warren <swa...@nvidia.com>
---
arch/arm/configs/tegra_defconfig | 1 +
1 file changed, 1 insertion(+)

diff --git a/arch/arm/configs/tegra_defconfig b/arch/arm/configs/tegra_defconfig
index ea042e8..dddb97b 100644
--- a/arch/arm/configs/tegra_defconfig
+++ b/arch/arm/configs/tegra_defconfig
@@ -32,6 +32,7 @@ CONFIG_PCI=y
CONFIG_PCI_MSI=y
CONFIG_PCI_TEGRA=y
CONFIG_PCIEPORTBUS=y
+CONFIG_TRUSTED_FOUNDATIONS=y
CONFIG_SMP=y
CONFIG_PREEMPT=y
CONFIG_AEABI=y

Stephen Warren

unread,
Oct 14, 2013, 6:20:01 PM10/14/13
to
On 10/11/2013 03:45 PM, Alexandre Courbot wrote:
> Trusted Foundations is a TrustZone-based secure monitor for ARM that
> can be invoked using the same SMC-based API on all supported
> platforms. This patch adds initial basic support for Trusted
> Foundations using the ARM firmware API. Current features are limited
> to the ability to boot secondary processors.
>
> Note: The API followed by Trusted Foundations does *not* follow the SMC
> calling conventions. It has nothing to do with PSCI neither and is only
> relevant to devices that use Trusted Foundations (like most Tegra-based
> retail devices).

Russell, are you OK with this patch? Is it OK if I take it through the
Tegra tree, or would you like to take it through your tree and give me a
stable branch I can apply the rest on top of?

Thanks.

Russell King - ARM Linux

unread,
Oct 15, 2013, 7:10:02 AM10/15/13
to
Is having this under arch/arm appropriate? What happens if the API
gets re-used on ARM64 for example? Would drivers/firmware be a better
cross-arch location for this?
Please move these up along side the other #include - having includes depend
on config symbols is an additional unnecessary source of fragility.

Secondly, please use linux/*.h includes in preference to asm/*.h where the
linux/*.h include picks up the corresponding asm/*.h include. In this case,
that should be linux/bug.h, not asm/bug.h.

Stephen Warren

unread,
Oct 15, 2013, 12:20:01 PM10/15/13
to
That's probably a better location, yes.

BTW, drivers/firmware/ doesn't seem to have an overall maintainer. Who
should Alex send the patch to, or should I just take this patch throught
the Tegra tree once the files are moved?

Alex Courbot

unread,
Oct 15, 2013, 1:30:01 PM10/15/13
to
The reason why this has been put into arch/arm is that the firmware_ops
feature this patch depends also resides there
(arch/arm/include/asm/firmware.h).

On the other hand it might also make sense to move firmware_ops out of
ARM since I don't see anything ARM-specific with it. Tomasz, could we
have your thoughts on this?
Will do, thanks!

Alex.

Tomasz Figa

unread,
Oct 28, 2013, 7:10:02 AM10/28/13
to
Hi,

2013/10/15 Alex Courbot <acou...@nvidia.com>:
I don't see anything wrong in moving this out of arch/arm, feel free
to do so.

However I guess that some (or all) of the names will have to
be put into a more separate namespace, as the term "firmware" is a bit
too generic IMHO. Possibly something like platform_firmware could be
better.

What do you think?

Best regards,
Tomasz

Olof Johansson

unread,
Oct 28, 2013, 6:00:02 PM10/28/13
to
I think we can probably merge this under arch/arm now, and when we figure out
what needs to be common with ARM64 we can move it out to a good location. It
might be that mostly just a header file with ABI conventions needs to be
shared, not actual implementation, for example.


-Olof
0 new messages