vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

PATCH New Sudeep Holla 4 participants 13 messages Sep 23, 2016 – Oct 8, 2016 Arm/versatile express platform

Share & tools

Sudeep Holla

author

2016-09-23 15:10 UTC #0 lore raw

PATCH 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.

However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.

Cc Jon Medhurst <[email protected]>

Signed-off-by Sudeep Holla <[email protected]>

---
 board/armltd/vexpress/vexpress_tc2.c | 47 ++++++++++++++++++++++++++++++++++++
 configs/vexpress_ca15_tc2_defconfig  |  1 +
 2 files changed, 48 insertions(+)

Hi,

This change is needed to avoid the kernel panic while attempting to access CCI ports when booting in non-sec/HYP mode. The kernel patches to fix this are available @[1]

Regards,
Sudeep

[1] http://www.spinics.net/lists/arm-kernel/msg533715.html

board/armltd/vexpress/vexpress_tc2.c +47 -0

index ebb41a8833ab..25bbba7a8590 100644
@@ -8,6 +8,9 @@
  */
 
 #include <asm/io.h>
+#include <asm/u-boot.h>
+#include <common.h>
+#include <libfdt.h>
 
 #define SCC_BASE	0x7fff0000

@@ -31,3 +34,47 @@ bool armv7_boot_nonsec_default(void)
 	return (readl((u32 *)(SCC_BASE + 0x700)) & ((1 << 12) | (1 << 13))) == 0;
 #endif
 }
+
+#ifdef CONFIG_OF_BOARD_SETUP
+int ft_board_setup(void *fdt, bd_t *bd)
+{
+	int offset, tmp, len;
+	const struct fdt_property *prop;
+	const char *cci_compatible = "arm,cci-400-ctrl-if";
+
+	if (!armv7_boot_nonsec_default())
+		return 0; /* Do nothing */
+
+	offset = fdt_path_offset(fdt, "/cpus");
+	if (offset < 0) {
+		printf("couldn't find /cpus\n");
+		return offset;
+	}
+
+	/* delete cci-control-port in each cpu node */
+	for (tmp = fdt_first_subnode(fdt, offset); tmp >= 0;
+	     tmp = fdt_next_subnode(fdt, tmp))
+		fdt_delprop(fdt, tmp, "cci-control-port");
+
+	/* disable all ace cci slave ports */
+	offset = fdt_node_offset_by_prop_value(fdt, offset, "compatible",
+					       cci_compatible, 20);
+	while (offset > 0) {
+		prop = fdt_get_property(fdt, offset, "interface-type",
+					&len);
+		if (!prop)
+			continue;
+		if (len < 4)
+			continue;
+		if (strcmp(prop->data, "ace"))
+			continue;
+
+		fdt_setprop_string(fdt, offset, "status", "disabled");
+
+		offset = fdt_node_offset_by_prop_value(fdt, offset, "compatible",
+						       cci_compatible, 20);
+	}
+
+	return 0;
+}
+#endif /* CONFIG_OF_BOARD_SETUP */

configs/vexpress_ca15_tc2_defconfig +1 -0

index 2f141dda06c6..5154803b7c65 100644
@@ -1,5 +1,6 @@
 CONFIG_ARM=y
 CONFIG_TARGET_VEXPRESS_CA15_TC2=y
+CONFIG_OF_BOARD_SETUP=y
 CONFIG_HUSH_PARSER=y
 # CONFIG_CMD_CONSOLE is not set
 # CONFIG_CMD_BOOTD is not set

--
2.7.4

Marc Zyngier

2016-09-23 15:27 UTC #1 lore raw

PATCH 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Quoted 11 lines · Sudeep Holla

Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.
Cc Jon Medhurst <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>

Ah, that'd be a great reason for me to update the 2.5 year old u-boot that lives on my TC2!

Acked-by Marc Zyngier <[email protected]>

M.

Signature · 1 line

Jazz is not dead. It just smells funny...

Jon Medhurst (Tixy)

2016-09-23 16:03 UTC #2 lore raw

PATCH 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

In reply to Sudeep Holla

board/armltd/vexpress/vexpress_tc2.c bool armv7_boot_nonsec_default(void) · line 47

Quoted 10 lines · Sudeep Holla

board/armltd/vexpress/vexpress_tc2.c +10 -0

+#ifdef CONFIG_OF_BOARD_SETUP
+int ft_board_setup(void *fdt, bd_t *bd)
+{
+       int offset, tmp, len;
+       const struct fdt_property *prop;
+       const char *cci_compatible = "arm,cci-400-ctrl-if";
+
+       if (!armv7_boot_nonsec_default())
+               return 0; /* Do nothing */
+

That's just testing the default not whether the board is actually going to boot in nonsec (hyp) mode or not, which also also depends on environment variables. So you probably want instead to call armv7_boot_nonsec() here. That function only exists if CONFIG_ARMV7_NONSEC is defined, so the 'if' statement above probably wants to be something like...

#ifdef CONFIG_ARMV7_NONSEC
	if (!armv7_boot_nonsec())
		return 0;
#else
	return 0;
#endif
	/* We only get here if board will boot in nonsec mode */

Signature · 1 line

Tixy

Sudeep Holla

author

2016-09-23 16:13 UTC #3 lore raw

PATCH 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Quoted 15 lines · Jon Medhurst (Tixy)

board/armltd/vexpress/vexpress_tc2.c bool armv7_boot_nonsec_default(void) · line 47
Quoted 10 lines · Sudeep Holla
board/armltd/vexpress/vexpress_tc2.c +10 -0
+#ifdef CONFIG_OF_BOARD_SETUP
+int ft_board_setup(void *fdt, bd_t *bd)
+{
+       int offset, tmp, len;
+       const struct fdt_property *prop;
+       const char *cci_compatible = "arm,cci-400-ctrl-if";
+
+       if (!armv7_boot_nonsec_default())
+               return 0; /* Do nothing */
+
That's just testing the default not whether the board is actually going to boot in nonsec (hyp) mode or not, which also also depends on environment variables.

Ah ok, wasn't aware of that. Thanks for pointing that out.

Quoted 5 lines

So you probably want instead to call
armv7_boot_nonsec() here. That function only exists if CONFIG_ARMV7_NONSEC is defined, so the 'if' statement above probably wants to be something like...

Quoted 8 lines

#ifdef CONFIG_ARMV7_NONSEC
	if (!armv7_boot_nonsec())
		return 0;
#else
	return 0;
#endif
	/* We only get here if board will boot in nonsec mode */

OK, will update accordingly.

Signature · 2 lines

Regards,
Sudeep

Sudeep Holla

author

2016-09-23 16:38 UTC #4 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

In reply to Sudeep Holla

Cc Jon Medhurst <[email protected]>

Acked-by Marc Zyngier <[email protected]>

Signed-off-by Sudeep Holla <[email protected]>

---
 board/armltd/vexpress/vexpress_tc2.c | 52 ++++++++++++++++++++++++++++++++++++
 configs/vexpress_ca15_tc2_defconfig  |  1 +
 2 files changed, 53 insertions(+)

Hi,

This change is needed to avoid the kernel panic while attempting to access CCI ports when booting in non-sec/HYP mode. The kernel patches to fix this are available @[1]

Regards,
Sudeep

v1->v2:
- Fix compilation with !CONFIG_ARMV7_NONSEC(Thanks to Tixy)

[1] http://www.spinics.net/lists/arm-kernel/msg533715.html

board/armltd/vexpress/vexpress_tc2.c +52 -0

index ebb41a8833ab..c7adf950f579 100644
@@ -7,7 +7,11 @@
  * SPDX-License-Identifier:	GPL-2.0+
  */
 
+#include <asm/armv7.h>
 #include <asm/io.h>
+#include <asm/u-boot.h>
+#include <common.h>
+#include <libfdt.h>
 
 #define SCC_BASE	0x7fff0000

@@ -31,3 +35,51 @@ bool armv7_boot_nonsec_default(void)
 	return (readl((u32 *)(SCC_BASE + 0x700)) & ((1 << 12) | (1 << 13))) == 0;
 #endif
 }
+
+#ifdef CONFIG_OF_BOARD_SETUP
+int ft_board_setup(void *fdt, bd_t *bd)
+{
+	int offset, tmp, len;
+	const struct fdt_property *prop;
+	const char *cci_compatible = "arm,cci-400-ctrl-if";
+
+#ifdef CONFIG_ARMV7_NONSEC
+	if (!armv7_boot_nonsec())
+		return 0;
+#else
+	return 0;
+#endif
+	/* Booting in nonsec mode, disable CCI access */
+	offset = fdt_path_offset(fdt, "/cpus");
+	if (offset < 0) {
+		printf("couldn't find /cpus\n");
+		return offset;
+	}
+
+	/* delete cci-control-port in each cpu node */
+	for (tmp = fdt_first_subnode(fdt, offset); tmp >= 0;
+	     tmp = fdt_next_subnode(fdt, tmp))
+		fdt_delprop(fdt, tmp, "cci-control-port");
+
+	/* disable all ace cci slave ports */
+	offset = fdt_node_offset_by_prop_value(fdt, offset, "compatible",
+					       cci_compatible, 20);
+	while (offset > 0) {
+		prop = fdt_get_property(fdt, offset, "interface-type",
+					&len);
+		if (!prop)
+			continue;
+		if (len < 4)
+			continue;
+		if (strcmp(prop->data, "ace"))
+			continue;
+
+		fdt_setprop_string(fdt, offset, "status", "disabled");
+
+		offset = fdt_node_offset_by_prop_value(fdt, offset, "compatible",
+						       cci_compatible, 20);
+	}
+
+	return 0;
+}
+#endif /* CONFIG_OF_BOARD_SETUP */

configs/vexpress_ca15_tc2_defconfig +1 -0

index 2f141dda06c6..5154803b7c65 100644
@@ -1,5 +1,6 @@
 CONFIG_ARM=y
 CONFIG_TARGET_VEXPRESS_CA15_TC2=y
+CONFIG_OF_BOARD_SETUP=y
 CONFIG_HUSH_PARSER=y
 # CONFIG_CMD_CONSOLE is not set
 # CONFIG_CMD_BOOTD is not set

--
2.7.4

Jon Medhurst (Tixy)

2016-09-26 11:30 UTC #5 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Quoted 9 lines · Sudeep Holla

Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.

Quoted 4 lines

Cc Jon Medhurst <[email protected]>
Acked-by Marc Zyngier <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>
---

This works for me (unsurprisingly) when booting in secure mode. What kernel and firmware config do I need to use for non-sec mode? I tried vexpress_defconfig, with bits 12 and 13 cleared in SCC: 0x700 but I get nothing out the console after "Starting kernel ..."

Signature · 1 line

Tixy

Sudeep Holla

author

2016-09-26 11:35 UTC #6 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Quoted 15 lines · Jon Medhurst (Tixy)

Quoted 9 lines · Sudeep Holla
Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.
Quoted 4 lines
Cc Jon Medhurst <[email protected]>
Acked-by Marc Zyngier <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>
---

Quoted 5 lines

This works for me (unsurprisingly) when booting in secure mode. What kernel and firmware config do I need to use for non-sec mode? I tried vexpress_defconfig, with bits 12 and 13 cleared in SCC: 0x700 but I get nothing out the console after "Starting kernel ..."

I just flipped bit 12 and 13 in SCC: 0x700 to switch between MCPM/secure and HYP/non-secure mode. All other images/settings remain the same.

So IIUC, with vexpress_defconfig + above SCC change you are seeing a hand, I will check that. I am using multi_v7_defconfig.

Signature · 2 lines

Regards,
Sudeep

Sudeep Holla

author

2016-09-26 11:37 UTC #7 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Quoted 25 lines · Sudeep Holla

Quoted 16 lines · Jon Medhurst (Tixy)
Quoted 10 lines · Sudeep Holla
Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting
to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.
Quoted 4 lines
Cc Jon Medhurst <[email protected]>
Acked-by Marc Zyngier <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>
---
Quoted 5 lines
This works for me (unsurprisingly) when booting in secure mode. What kernel and firmware config do I need to use for non-sec mode? I tried vexpress_defconfig, with bits 12 and 13 cleared in SCC: 0x700 but I get nothing out the console after "Starting kernel ..."

Quoted 6 lines

I just flipped bit 12 and 13 in SCC: 0x700 to switch between MCPM/secure and HYP/non-secure mode. All other images/settings remain the same.
So IIUC, with vexpress_defconfig + above SCC change you are seeing a hand, I will check that. I am using multi_v7_defconfig.

+ the patches from Lorenzo fixing MCPM code in the kernel [1] as mentioned first email carrying this patch.

Signature · 4 lines

Regards,
Sudeep

[1] http://www.spinics.net/lists/arm-kernel/msg533715.html

Jon Medhurst (Tixy)

2016-09-26 12:30 UTC #8 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

In reply to Sudeep Holla

Quoted 9 lines · Sudeep Holla

Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.

Quoted 4 lines

Cc Jon Medhurst <[email protected]>
Acked-by Marc Zyngier <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>
---

Acked-by Jon Medhurst <[email protected]>

Tested-by Jon Medhurst <[email protected]>

Quoted 100 lines

 board/armltd/vexpress/vexpress_tc2.c | 52 ++++++++++++++++++++++++++++++++++++
 configs/vexpress_ca15_tc2_defconfig  |  1 +
 2 files changed, 53 insertions(+)

Hi,

This change is needed to avoid the kernel panic while attempting to access CCI ports when booting in non-sec/HYP mode. The kernel patches to fix this are available @[1]

Regards,
Sudeep

v1->v2:
- Fix compilation with !CONFIG_ARMV7_NONSEC(Thanks to Tixy)

[1] http://www.spinics.net/lists/arm-kernel/msg533715.html

board/armltd/vexpress/vexpress_tc2.c +52 -0

index ebb41a8833ab..c7adf950f579 100644
@@ -7,7 +7,11 @@
  * SPDX-License-Identifier:	GPL-2.0+
  */
 
+#include <asm/armv7.h>
 #include <asm/io.h>
+#include <asm/u-boot.h>
+#include <common.h>
+#include <libfdt.h>
 
 #define SCC_BASE	0x7fff0000

@@ -31,3 +35,51 @@ bool armv7_boot_nonsec_default(void)
 	return (readl((u32 *)(SCC_BASE + 0x700)) & ((1 << 12) | (1 << 13))) == 0;
 #endif
 }
+
+#ifdef CONFIG_OF_BOARD_SETUP
+int ft_board_setup(void *fdt, bd_t *bd)
+{
+	int offset, tmp, len;
+	const struct fdt_property *prop;
+	const char *cci_compatible = "arm,cci-400-ctrl-if";
+
+#ifdef CONFIG_ARMV7_NONSEC
+	if (!armv7_boot_nonsec())
+		return 0;
+#else
+	return 0;
+#endif
+	/* Booting in nonsec mode, disable CCI access */
+	offset = fdt_path_offset(fdt, "/cpus");
+	if (offset < 0) {
+		printf("couldn't find /cpus\n");
+		return offset;
+	}
+
+	/* delete cci-control-port in each cpu node */
+	for (tmp = fdt_first_subnode(fdt, offset); tmp >= 0;
+	     tmp = fdt_next_subnode(fdt, tmp))
+		fdt_delprop(fdt, tmp, "cci-control-port");
+
+	/* disable all ace cci slave ports */
+	offset = fdt_node_offset_by_prop_value(fdt, offset, "compatible",
+					       cci_compatible, 20);
+	while (offset > 0) {
+		prop = fdt_get_property(fdt, offset, "interface-type",
+					&len);
+		if (!prop)
+			continue;
+		if (len < 4)
+			continue;
+		if (strcmp(prop->data, "ace"))
+			continue;
+
+		fdt_setprop_string(fdt, offset, "status", "disabled");
+
+		offset = fdt_node_offset_by_prop_value(fdt, offset, "compatible",
+						       cci_compatible, 20);
+	}
+
+	return 0;
+}
+#endif /* CONFIG_OF_BOARD_SETUP */

configs/vexpress_ca15_tc2_defconfig +1 -1

index 2f141dda06c6..5154803b7c65 100644
@@ -1,5 +1,6 @@
 CONFIG_ARM=y
 CONFIG_TARGET_VEXPRESS_CA15_TC2=y
+CONFIG_OF_BOARD_SETUP=y
 CONFIG_HUSH_PARSER=y
 # CONFIG_CMD_CONSOLE is not set
 # CONFIG_CMD_BOOTD is not set
--

2.7.4

Sudeep Holla

author

2016-09-26 12:38 UTC #9 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Quoted 10 lines · Jon Medhurst (Tixy)

Quoted 9 lines · Sudeep Holla
Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.

Quoted 7 lines

Quoted 4 lines
Cc Jon Medhurst <[email protected]>
Acked-by Marc Zyngier <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>
---
Acked-by Jon Medhurst <[email protected]>
Tested-by Jon Medhurst <[email protected]>

So, can I assume the missing kernel patches to be reason for boot hang ? Just wanted to know if I need to investigate that any further ?

Signature · 2 lines

Regards,
Sudeep

Jon Medhurst (Tixy)

2016-09-26 13:19 UTC #10 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Quoted 17 lines · Sudeep Holla

Quoted 15 lines · Jon Medhurst (Tixy)
Quoted 13 lines · Sudeep Holla
Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.
Cc Jon Medhurst <[email protected]>
Acked-by Marc Zyngier <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>
---

Quoted 5 lines

Quoted 2 lines
Acked-by Jon Medhurst <[email protected]>
Tested-by Jon Medhurst <[email protected]>
So, can I assume the missing kernel patches to be reason for boot hang ? Just wanted to know if I need to investigate that any further ?

Sorry, yes they were the reason and no further investigation needed. I remembered getting nonsec mode working some month's ago without such patches, but I remember now that was by disabling MCPM in the kernel.

This morning I tried these U-Boot patches successfully with:
- Upstream vexpress_defconfig kernel booting with 'sec' mode
- That kernel with Lorenzo's patches in both 'sec' and 'nonsec'
- As above with CONFIG_BL_SWITCHER enabled

When booting in nonsec I also verified the device-tree modifications made by this patch by seeing the following files existed and contained the word 'disabled'...

/proc/device-tree/cci at 2c090000/slave-if at 4000/status /proc/device-tree/cci at 2c090000/slave-if at 5000/status

Signature · 1 line

Tixy

Sudeep Holla

author

2016-09-26 13:24 UTC #11 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

Quoted 33 lines · Jon Medhurst (Tixy)

Quoted 17 lines · Sudeep Holla
Quoted 15 lines · Jon Medhurst (Tixy)
Quoted 13 lines · Sudeep Holla
Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.
Cc Jon Medhurst <[email protected]>
Acked-by Marc Zyngier <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>
---
Quoted 5 lines
Quoted 2 lines
Acked-by Jon Medhurst <[email protected]>
Tested-by Jon Medhurst <[email protected]>
So, can I assume the missing kernel patches to be reason for boot hang ? Just wanted to know if I need to investigate that any further ?
Sorry, yes they were the reason and no further investigation needed. I remembered getting nonsec mode working some month's ago without such patches, but I remember now that was by disabling MCPM in the kernel.
This morning I tried these U-Boot patches successfully with:
- Upstream vexpress_defconfig kernel booting with 'sec' mode
- That kernel with Lorenzo's patches in both 'sec' and 'nonsec'
- As above with CONFIG_BL_SWITCHER enabled

Quoted 7 lines

When booting in nonsec I also verified the device-tree modifications made by this patch by seeing the following files existed and contained the word 'disabled'...
/proc/device-tree/cci at 2c090000/slave-if at 4000/status /proc/device-tree/cci at 2c090000/slave-if at 5000/status

That was very detailed :), thanks for testing and confirming. I too did similar examination and didn't find anything odd so far.

Signature · 2 lines

Regards,
Sudeep

Tom Rini

2016-10-08 17:06 UTC #12 lore raw

PATCH v2 1/1 vexpress: disable cci ace slave ports when booting in non-sec/hyp mode

In reply to Sudeep Holla

Quoted 9 lines · Sudeep Holla

Commit f225d39d3093 ("vexpress: Check TC2 firmware support before defaulting to nonsec booting") added support to check if the firmware on TC2 is configured appropriately before booting in nonsec/hyp mode.
However when booting in non-secure/hyp mode, CCI control must be done in secure firmware and can't be done in non-secure/hyp mode. In order to ensure that, this patch disables the cci slave port inteface so that it is not accessed at all.

Quoted 5 lines

Cc Jon Medhurst <[email protected]>
Acked-by Marc Zyngier <[email protected]>
Signed-off-by Sudeep Holla <[email protected]>
Acked-by Jon Medhurst <[email protected]>
Tested-by Jon Medhurst <[email protected]>

Applied to u-boot/master, thanks!

Signature · 8 lines

Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20161008/851f83d8/attachment.sig>