Skip to contentSkip to conversation

cm-fx6 updates

PATCH New Nikita Kiryanov 4 participants 16 messages Oct 29, 2014 – Nov 6, 2014

lore
Share & tools
lore

Nikita Kiryanov

author
#0 lore raw

v2 cover cm-fx6 updates

This patchset contains a bug fix for DRAM detection, support for Phison SSD, and a new preboot hook.

Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>

Changes in V2:
- Added board_preboot_os to bootm.h
- Split "common: introduce board_preboot_os hook" into 2 patches

Nikita Kiryanov (4):
arm: mx6: cm_fx6: change issd gpio order
arm: mx6: cm_fx6: detect 1GB DRAM correctly on solo
common: introduce board_preboot_os hook
arm: mx6: cm_fx6: power down sata on OS boot

 board/compulab/cm_fx6/cm_fx6.c | 8 +++++++-
 board/compulab/cm_fx6/spl.c    | 7 ++++---
 common/bootm_os.c              | 7 +++++++
 include/bootm.h                | 1 +
 4 files changed, 19 insertions(+), 4 deletions(-)
Signature · 1 line
1.9.1
author
#1 lore raw

PATCH v2 1/4 arm: mx6: cm_fx6: change issd gpio order

Change the order in which GPIOs are toggled in SATA init sequence to accomodate both SanDisk and Phison SSDs.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
---
 board/compulab/cm_fx6/cm_fx6.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
board/compulab/cm_fx6/cm_fx6.c +1 -1
index 82681b1..0206ae8 100644
@@ -31,12 +31,12 @@ DECLARE_GLOBAL_DATA_PTR;
 #ifdef CONFIG_DWC_AHSATA
 static int cm_fx6_issd_gpios[] = {
 	/* The order of the GPIOs in the array is important! */
+	CM_FX6_SATA_LDO_EN,
 	CM_FX6_SATA_PHY_SLP,
 	CM_FX6_SATA_NRSTDLY,
 	CM_FX6_SATA_PWREN,
 	CM_FX6_SATA_NSTANDBY1,
 	CM_FX6_SATA_NSTANDBY2,
-	CM_FX6_SATA_LDO_EN,
 };
 
 static void cm_fx6_sata_power(int on)
Signature · 1 line
1.9.1
author
#2 lore raw

PATCH v2 1/4 arm: mx6: cm_fx6: change issd gpio order

Forgot the changelog:

Quoted 7 lines · Nikita Kiryanov

Change the order in which GPIOs are toggled in SATA init sequence to accomodate both SanDisk and Phison SSDs.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>

---

Changes in V2:
- No changes.

Quoted 22 lines
  board/compulab/cm_fx6/cm_fx6.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
board/compulab/cm_fx6/cm_fx6.c +1 -1
index 82681b1..0206ae8 100644
@@ -31,12 +31,12 @@ DECLARE_GLOBAL_DATA_PTR;
  #ifdef CONFIG_DWC_AHSATA
  static int cm_fx6_issd_gpios[] = {
  	/* The order of the GPIOs in the array is important! */
+	CM_FX6_SATA_LDO_EN,
  	CM_FX6_SATA_PHY_SLP,
  	CM_FX6_SATA_NRSTDLY,
  	CM_FX6_SATA_PWREN,
  	CM_FX6_SATA_NSTANDBY1,
  	CM_FX6_SATA_NSTANDBY2,
-	CM_FX6_SATA_LDO_EN,
  };
 
  static void cm_fx6_sata_power(int on)
Signature · 2 lines
Regards,
Nikita Kiryanov
author
#4 lore raw

PATCH v2 2/4 arm: mx6: cm_fx6: detect 1GB DRAM correctly on solo

In reply to Nikita Kiryanov

The 1GB DRAM configuration on mx6 solo uses 2 chip selects, but the code tests 1GB DRAM configuration as if it is all present on one chip select, and thus cannot see the full range of available memory.

Refactor the check to detect 1GB DRAM correctly.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
---
 board/compulab/cm_fx6/spl.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
board/compulab/cm_fx6/spl.c +4 -3
index 3948ba2..6fe937b 100644
@@ -235,10 +235,11 @@ static int cm_fx6_spl_dram_init(void)
 
 		spl_mx6s_dram_init(DDR_32BIT_1GB, false);
 		bank1_size = get_ram_size((long int *)PHYS_SDRAM_1, 0x80000000);
-		if (bank1_size == 0x40000000)
-			return 0;
-
+		bank2_size = get_ram_size((long int *)PHYS_SDRAM_2, 0x80000000);
 		if (bank1_size == 0x20000000) {
+			if (bank2_size == 0x20000000)
+				return 0;
+
 			spl_mx6s_dram_init(DDR_32BIT_512MB, true);
 			return 0;
 		}
Signature · 1 line
1.9.1
author
#5 lore raw

PATCH v2 2/4 arm: mx6: cm_fx6: detect 1GB DRAM correctly on solo

Forgot the changelog:

Quoted 10 lines · Nikita Kiryanov

The 1GB DRAM configuration on mx6 solo uses 2 chip selects, but the code tests 1GB DRAM configuration as if it is all present on one chip select, and thus cannot see the full range of available memory.

Refactor the check to detect 1GB DRAM correctly.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>

---

Changes in V2:
- No changes.

Quoted 23 lines
  board/compulab/cm_fx6/spl.c | 7 ++++---
  1 file changed, 4 insertions(+), 3 deletions(-)
board/compulab/cm_fx6/spl.c +4 -3
index 3948ba2..6fe937b 100644
@@ -235,10 +235,11 @@ static int cm_fx6_spl_dram_init(void)
 
  		spl_mx6s_dram_init(DDR_32BIT_1GB, false);
  		bank1_size = get_ram_size((long int *)PHYS_SDRAM_1, 0x80000000);
-		if (bank1_size == 0x40000000)
-			return 0;
-
+		bank2_size = get_ram_size((long int *)PHYS_SDRAM_2, 0x80000000);
  		if (bank1_size == 0x20000000) {
+			if (bank2_size == 0x20000000)
+				return 0;
+
  			spl_mx6s_dram_init(DDR_32BIT_512MB, true);
  			return 0;
  		}
Signature · 2 lines
Regards,
Nikita Kiryanov
#6 lore raw

PATCH v2 2/4 arm: mx6: cm_fx6: detect 1GB DRAM correctly on solo

In reply to Nikita Kiryanov
Quoted 9 lines · Nikita Kiryanov

The 1GB DRAM configuration on mx6 solo uses 2 chip selects, but the code tests 1GB DRAM configuration as if it is all present on one chip select, and thus cannot see the full range of available memory.

Refactor the check to detect 1GB DRAM correctly.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Acked-by Igor Grinberg <[email protected]>
Signature · 2 lines
Regards,
Igor.
author
#7 lore raw

PATCH v2 3/4 common: introduce board_preboot_os hook

In reply to Nikita Kiryanov

Introduce board specific function board_preboot_os() to allow for board specific config before we boot.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Cc Tom Rini <[email protected]>
Cc Jeroen Hofstee <[email protected]>
Cc Otavio Salvador <[email protected]>

---
Changes in V2:
- Added board_preboot_os to bootm.h
- Split cm_fx6 stuff into a separate patch

 common/bootm_os.c | 7 +++++++
 include/bootm.h   | 1 +
 2 files changed, 8 insertions(+)
common/bootm_os.c +7 -0
index 5be4467..95cd657 100644
@@ -442,10 +442,17 @@ __weak void arch_preboot_os(void)
 	/* please define platform specific arch_preboot_os() */
 }
 
+/* Allow for board specific config before we boot */
+__weak void board_preboot_os(void)
+{
+	/* please define board specific board_preboot_os() */
+}
+
 int boot_selected_os(int argc, char * const argv[], int state,
 		     bootm_headers_t *images, boot_os_fn *boot_fn)
 {
 	arch_preboot_os();
+	board_preboot_os();
 	boot_fn(state, argc, argv, images);
 
 	/* Stand-alone may return when 'autostart' is 'no' */
include/bootm.h +1 -0
index b3d1a62..7a57264 100644
@@ -55,5 +55,6 @@ int do_bootm_states(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[],
 		    int states, bootm_headers_t *images, int boot_progress);
 
 void arch_preboot_os(void);
+void board_preboot_os(void);
 
 #endif
Signature · 1 line
1.9.1
#8 lore raw

PATCH v2 3/4 common: introduce board_preboot_os hook

Quoted 9 lines · Nikita Kiryanov

Introduce board specific function board_preboot_os() to allow for board specific config before we boot.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Cc Tom Rini <[email protected]>
Cc Jeroen Hofstee <[email protected]>
Cc Otavio Salvador <[email protected]>
Acked-by Otavio Salvador <[email protected]>
Signature · 3 lines
Otavio Salvador                             O.S. Systems
http://www.ossystems.com.br        http://code.ossystems.com.br
Mobile: +55 (53) 9981-7854            Mobile: +1 (347) 903-9750
#9 lore raw

PATCH v2 3/4 common: introduce board_preboot_os hook

In reply to Nikita Kiryanov
Quoted 9 lines · Nikita Kiryanov

Introduce board specific function board_preboot_os() to allow for board specific config before we boot.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Cc Tom Rini <[email protected]>
Cc Jeroen Hofstee <[email protected]>
Cc Otavio Salvador <[email protected]>
Acked-by Igor Grinberg <[email protected]>
Signature · 2 lines
Regards,
Igor.
#10 lore raw

PATCH v2 3/4 common: introduce board_preboot_os hook

In reply to Nikita Kiryanov

Hi Nikita,

Quoted 9 lines · Nikita Kiryanov

Introduce board specific function board_preboot_os() to allow for board specific config before we boot.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Cc Tom Rini <[email protected]>
Cc Jeroen Hofstee <[email protected]>
Cc Otavio Salvador <[email protected]>
Quoted 9 lines

---
Changes in V2:
- Added board_preboot_os to bootm.h
- Split cm_fx6 stuff into a separate patch

 common/bootm_os.c | 7 +++++++
 include/bootm.h   | 1 +
 2 files changed, 8 insertions(+)

There is something that does not convince me. Really, the general statement should be to turn all devices off before booting the kernel, because this is what Linux expects. That said, this could be later solved with dm, because we will can iterate through all drivers and call a shutdown function.

Currently, "preboot" functions are turning off the hardware, so the name is already misleading. And I see that arch_preboot_os() was also convinced (because it is weak) to become a board_preboot, for example
here: board/renesas/koelsch/koelsch.c (and in other renesas implementation, too).

I have some concerns adding a new weak function, that enables some further hooks inside board code. I find it not well scalable. In your case, you need such as sata_shutdown(), and it should be responsibility of the general sata code to call something specific for the board, if necessary.

In current code, bootm_disable_interrupts() calls also usb_stop() and eth_halt(), - nothing to do with the name of the function. IMHO it should be better to move the code to shutdown interface inside boot_selected_os(), or having a common function "disable_hardware()" that calls usb_stop(), eth_halt() and then, why not, a sata_stop().

There is also in arch/arm/imx-common/cpu.c a derived implementation for arch_preboot_os(). Really it has nothing to do with arch, as I can see:
it shuts down only the IPU. But it is another hook, and IMHO it is better stopping sata here as adding a hidden callback.

Best regards,
Stefano Babic

Quoted 34 lines
common/bootm_os.c +7 -0
index 5be4467..95cd657 100644
@@ -442,10 +442,17 @@ __weak void arch_preboot_os(void)
 	/* please define platform specific arch_preboot_os() */
 }
 
+/* Allow for board specific config before we boot */
+__weak void board_preboot_os(void)
+{
+	/* please define board specific board_preboot_os() */
+}
+
 int boot_selected_os(int argc, char * const argv[], int state,
 		     bootm_headers_t *images, boot_os_fn *boot_fn)
 {
 	arch_preboot_os();
+	board_preboot_os();
 	boot_fn(state, argc, argv, images);
 
 	/* Stand-alone may return when 'autostart' is 'no' */
include/bootm.h +1 -0
index b3d1a62..7a57264 100644
@@ -55,5 +55,6 @@ int do_bootm_states(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[],
 		    int states, bootm_headers_t *images, int boot_progress);
 
 void arch_preboot_os(void);
+void board_preboot_os(void);
 
 #endif
Signature · 5 lines
=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sbabic at denx.de
=====================================================================
author
#11 lore raw

PATCH v2 3/4 common: introduce board_preboot_os hook

Hi Stefano,

Quoted 46 lines · Stefano Babic

Hi Nikita,

Quoted 9 lines · Nikita Kiryanov

Introduce board specific function board_preboot_os() to allow for board specific config before we boot.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Cc Tom Rini <[email protected]>
Cc Jeroen Hofstee <[email protected]>
Cc Otavio Salvador <[email protected]>
Quoted 9 lines

---
Changes in V2:
- Added board_preboot_os to bootm.h
- Split cm_fx6 stuff into a separate patch

  common/bootm_os.c | 7 +++++++
  include/bootm.h   | 1 +
  2 files changed, 8 insertions(+)

There is something that does not convince me. Really, the general statement should be to turn all devices off before booting the kernel, because this is what Linux expects. That said, this could be later solved with dm, because we will can iterate through all drivers and call a shutdown function.

Currently, "preboot" functions are turning off the hardware, so the name is already misleading. And I see that arch_preboot_os() was also convinced (because it is weak) to become a board_preboot, for example
here: board/renesas/koelsch/koelsch.c (and in other renesas implementation, too).

I have some concerns adding a new weak function, that enables some further hooks inside board code. I find it not well scalable. In your case, you need such as sata_shutdown(), and it should be responsibility of the general sata code to call something specific for the board, if necessary.

In current code, bootm_disable_interrupts() calls also usb_stop() and eth_halt(), - nothing to do with the name of the function. IMHO it should be better to move the code to shutdown interface inside boot_selected_os(), or having a common function "disable_hardware()" that calls usb_stop(), eth_halt() and then, why not, a sata_stop().

Quoted 4 lines

There is also in arch/arm/imx-common/cpu.c a derived implementation for arch_preboot_os(). Really it has nothing to do with arch, as I can see:
it shuts down only the IPU. But it is another hook, and IMHO it is better stopping sata here as adding a hidden callback.

You make some good points.
I'll switch to using arch_preboot_os until there's a better way of doing this.

Signature · 2 lines
Regards,
Nikita Kiryanov
author
#12 lore raw

PATCH v2 4/4 arm: mx6: cm_fx6: power down sata on OS boot

In reply to Nikita Kiryanov

If sata is used by U-Boot, the Linux kernel fails to detect the ssd correctly afterwards. Power off sata on OS boot so that Linux will have a clean state to work with.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Cc Jeroen Hofstee <[email protected]>
Cc Otavio Salvador <[email protected]>

---
Changes in V2:
- New patch split off the previous patch.

 board/compulab/cm_fx6/cm_fx6.c | 6 ++++++
 1 file changed, 6 insertions(+)
board/compulab/cm_fx6/cm_fx6.c +6 -0
index 0206ae8..676de40 100644
@@ -15,6 +15,7 @@
 #include <netdev.h>
 #include <fdt_support.h>
 #include <sata.h>
+#include <bootm.h>
 #include <asm/arch/crm_regs.h>
 #include <asm/arch/sys_proto.h>
 #include <asm/arch/iomux.h>

@@ -125,6 +126,11 @@ int sata_initialize(void)
 
 	return err;
 }
+
+void board_preboot_os(void)
+{
+	cm_fx6_sata_power(0);
+}
 #else
 static int cm_fx6_setup_issd(void) { return 0; }
 #endif
Signature · 1 line
1.9.1
#13 lore raw

PATCH v2 4/4 arm: mx6: cm_fx6: power down sata on OS boot

Quoted 10 lines · Nikita Kiryanov

If sata is used by U-Boot, the Linux kernel fails to detect the ssd correctly afterwards. Power off sata on OS boot so that Linux will have a clean state to work with.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Cc Jeroen Hofstee <[email protected]>
Cc Otavio Salvador <[email protected]>

---

Acked-by Otavio Salvador <[email protected]>
Signature · 3 lines
Otavio Salvador                             O.S. Systems
http://www.ossystems.com.br        http://code.ossystems.com.br
Mobile: +55 (53) 9981-7854            Mobile: +1 (347) 903-9750
#14 lore raw

PATCH v2 4/4 arm: mx6: cm_fx6: power down sata on OS boot

In reply to Nikita Kiryanov
Quoted 9 lines · Nikita Kiryanov

If sata is used by U-Boot, the Linux kernel fails to detect the ssd correctly afterwards. Power off sata on OS boot so that Linux will have a clean state to work with.

Signed-off-by Nikita Kiryanov <[email protected]>
Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Cc Jeroen Hofstee <[email protected]>
Cc Otavio Salvador <[email protected]>
Acked-by Igor Grinberg <[email protected]>
Signature · 2 lines
Regards,
Igor.
#15 lore raw

v2 cover cm-fx6 updates

In reply to Nikita Kiryanov
Quoted 6 lines · Nikita Kiryanov

This patchset contains a bug fix for DRAM detection, support for Phison SSD, and a new preboot hook.

Cc Igor Grinberg <[email protected]>
Cc Stefano Babic <[email protected]>
Quoted 7 lines

Changes in V2:
- Added board_preboot_os to bootm.h
- Split "common: introduce board_preboot_os hook" into 2 patches

Nikita Kiryanov (4):
arm: mx6: cm_fx6: change issd gpio order
arm: mx6: cm_fx6: detect 1GB DRAM correctly on solo

First two patches applied to u-boot-imx, thanks !

Best regards,
Stefano Babic

Signature · 5 lines
=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sbabic at denx.de
=====================================================================