gmane.linux.hardware.blackfin.u-boot.devel

Re: [U-Boot] [PATCH v1 2/2] blackfin: bf609: add softswitch config command

Tom Rini — 2013-05-13T21:46:43

Yes, thanks.  And it looks fine, aside from not updating the README.

[PATCH v2 2/2] blackfin: bf609: add softswitchconfig command

Sonic Zhang — 2013-05-13T08:00:56

From: Bob Liu <lliubbo-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

Add softswitch_output command for bf609-ezkit to enable softswitches.

Signed-off-by: Bob Liu <lliubbo-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
Signed-off-by: Sonic Zhang <sonic.adi-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
---
Changes in v2:
- Add documentation for CONFIG_CMD_SOFTSWITCH to top-level README.

 README                                  |  1 +
 arch/blackfin/include/asm/soft_switch.h | 18 +++++++++++++++
 board/bf609-ezkit/soft_switch.c         | 11 +--------
 board/bf609-ezkit/soft_switch.h         | 25 +++++++++++++-------
 common/Makefile                         |  1 +
 common/cmd_softswitch.c                 | 41 +++++++++++++++++++++++++++++++++
 include/configs/bf609-ezkit.h           |  1 +
 7 files changed, 80 insertions(+), 18 deletions(-)
 create mode 100644 arch/blackfin/include/asm/soft_switch.h
 create mode 100644 common/cmd_softswitch.c

d

[PATCH v2 1/2] blackfin: bf609: implement soft switch

Sonic Zhang — 2013-05-13T08:00:55

From: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>

Set up soft switch pins properly in board init code.

Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
Signed-off-by: Scott Jiang <scott.jiang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
Signed-off-by: Bob Liu <lliubbo-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
Signed-off-by: Sonic Zhang <sonic.adi-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
---
Changes in v2: None

 board/bf609-ezkit/soft_switch.c | 180 ++++++++++++++++++++++++++++++++++++++++
 board/bf609-ezkit/soft_switch.h |  71 ++++++++++++++++
 2 files changed, 251 insertions(+)
 create mode 100644 board/bf609-ezkit/soft_switch.c
 create mode 100644 board/bf609-ezkit/soft_switch.h

diff --git a/board/bf609-ezkit/soft_switch.c b/board/bf609-ezkit/soft_switch.c
new file mode 100644
index 0000000..2e1404f
--- /dev/null
+++ b/board/bf609-ezkit/soft_switch.c
< at >< at > -0,0 +1,180 < at >< at >
+/*
+ * U-boot - main board file
+ *
+ * Copyright (c) 2008-2011 Analog Devices Inc.
+

[PATCH v2 2/2] blackfin: bf609: add softswitchconfig command

Sonic Zhang — 2013-05-13T07:59:27

From: Bob Liu <lliubbo-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

Add softswitch_output command for bf609-ezkit to enable softswitches.

Signed-off-by: Bob Liu <lliubbo-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
---
Changes in v2:
- Add documentation for CONFIG_CMD_SOFTSWITCH to top-level README.

 README                                  |  1 +
 arch/blackfin/include/asm/soft_switch.h | 18 +++++++++++++++
 board/bf609-ezkit/soft_switch.c         | 11 +--------
 board/bf609-ezkit/soft_switch.h         | 25 +++++++++++++-------
 common/Makefile                         |  1 +
 common/cmd_softswitch.c                 | 41 +++++++++++++++++++++++++++++++++
 include/configs/bf609-ezkit.h           |  1 +
 7 files changed, 80 insertions(+), 18 deletions(-)
 create mode 100644 arch/blackfin/include/asm/soft_switch.h
 create mode 100644 common/cmd_softswitch.c

diff --git a/README b/README
index 0d37d56..defdedb 100644
--- a/README
+++ b/RE

[PATCH v2 1/2] blackfin: bf609: implement soft switch

Sonic Zhang — 2013-05-13T07:59:26

Set up soft switch pins properly in board init code.

Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
Signed-off-by: Scott Jiang <scott.jiang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
Signed-off-by: Bob Liu <lliubbo-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
---
Changes in v2: None

 board/bf609-ezkit/soft_switch.c | 180 ++++++++++++++++++++++++++++++++++++++++
 board/bf609-ezkit/soft_switch.h |  71 ++++++++++++++++
 2 files changed, 251 insertions(+)
 create mode 100644 board/bf609-ezkit/soft_switch.c
 create mode 100644 board/bf609-ezkit/soft_switch.h

diff --git a/board/bf609-ezkit/soft_switch.c b/board/bf609-ezkit/soft_switch.c
new file mode 100644
index 0000000..2e1404f
--- /dev/null
+++ b/board/bf609-ezkit/soft_switch.c
< at >< at > -0,0 +1,180 < at >< at >
+/*
+ * U-boot - main board file
+ *
+ * Copyright (c) 2008-2011 Analog Devices Inc.
+ *
+ * Licensed under the GPL-2 or later.
+ */
+
+#include <common.h>
+#include <asm/blackfin.h>
+#include <asm/io.h>
+#include <i2c.h>
+#include "soft_s

Re: [PATCH v1 1/3] blackfin: Enable early print viathe generic serial API.

Mike Frysinger — 2013-05-13T04:42:11

the link address of the function doesn't matter.  all Blackfin func calls are 
pcrel based, not absolute address calls.  i specifically crafted all of this 
early serial code so that this does work regardless of the address where the 
code happens to be executing.

what doesn't work is if you try to use the early serial console with 
CONFIG_SYS_BFIN_UART (which allows all the serial devices to be selected on 
the fly).  but the header file specifically handles that:
#ifndef CONFIG_DEBUG_EARLY_SERIAL
# define CONFIG_SYS_BFIN_UART
#endif

i guess the problem is that this commit is wrong:

commit 820edc18b5aefa8c82d420f6cba3e13b5631f9b8
Author: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
Date:   Fri Nov 30 17:39:32 2012 +0800

    blackfin: Correct early serial mess output in BYPASS boot mode.

it shouldn't be messing with the addresses there
-mike
_______________________________________________
U-Boot-devel mailing list
U-Boot-devel-ZG0+EudsQA8dtHy/vicBwGD2FQJk+8+b< at >public.gmane.org
https

[PATCH v1 1/3] blackfin: Enable early print via thegeneric serial API.

Sonic Zhang — 2013-05-13T04:20:15

From: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>

Remove blackfin specific implementation of the generic serial API when
early print macro is defined.

In BFIN_BOOT_BYPASS mode, don't call generic serial_puts, because
early print in bypass mode is running before code binary is relocated
to the link address.

Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
---
 arch/blackfin/cpu/serial.c        |   71 ++++++++++++------------------------
 arch/blackfin/cpu/serial.h        |   10 +++---
 include/configs/bfin_adi_common.h |    3 --
 3 files changed, 29 insertions(+), 55 deletions(-)

diff --git a/arch/blackfin/cpu/serial.c b/arch/blackfin/cpu/serial.c
index 36d2a5c..89290d6 100644
--- a/arch/blackfin/cpu/serial.c
+++ b/arch/blackfin/cpu/serial.c
< at >< at > -197,13 +197,16 < at >< at > static void uart_loop(uint32_t uart_base, int state)
 
 static inline void __serial_set_baud(uint32_t uart_base, uint32_t baud)
 {
+#ifdef CONFIG_DEBUG_EARLY_SERIAL
+serial_early_set_baud(uar

[PATCH v1 3/3] blackfin: Uart divisor should be setafter their values are generated.

Sonic Zhang — 2013-05-13T04:20:17

From: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>

Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
---
 arch/blackfin/cpu/initcode.c |    7 +++++--
 1 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/arch/blackfin/cpu/initcode.c b/arch/blackfin/cpu/initcode.c
index 4b10b6c..8ef0b92 100644
--- a/arch/blackfin/cpu/initcode.c
+++ b/arch/blackfin/cpu/initcode.c
< at >< at > -715,8 +715,8 < at >< at > update_serial_clocks(ADI_BOOT_DATA *bs, uint sdivB, uint divB, uint vcoB)
  * for dividing which means we'd generate a libgcc reference.
  */
 unsigned int sdivR, vcoR;
-unsigned int dividend = sdivB * divB * vcoR;
-unsigned int divisor = vcoB * sdivR;
+unsigned int dividend;
+unsigned int divisor;
 unsigned int quotient;
 
 serial_putc('a');
< at >< at > -729,6 +729,9 < at >< at > update_serial_clocks(ADI_BOOT_DATA *bs, uint sdivB, uint divB, uint vcoB)
 sdivR = bfin_read_PLL_DIV() & 0xf;
 vcoR = (bfin_read_PLL_CTL() >> 9) & 0x3f;
 #endif
+
+dividend = sdivB * divB * vcoR;
+d

[PATCH v1 2/3] blackfin: Add memory virtual consoleto blackfin serial driver.

Sonic Zhang — 2013-05-13T04:20:16

From: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>

Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
---
 arch/blackfin/cpu/serial.c |   60 +++++++++++++++++++++++++++++++++++++++++++-
 1 files changed, 59 insertions(+), 1 deletions(-)

diff --git a/arch/blackfin/cpu/serial.c b/arch/blackfin/cpu/serial.c
index 89290d6..c453a03 100644
--- a/arch/blackfin/cpu/serial.c
+++ b/arch/blackfin/cpu/serial.c
< at >< at > -351,4 +351,62 < at >< at > void _serial_early_init(void)
 }
 #endif
 
-#endif
+#elif defined(CONFIG_UART_MEM)
+
+char serial_logbuf[CONFIG_UART_MEM];
+char *serial_logbuf_head = serial_logbuf;
+
+int serial_mem_init(void)
+{
+serial_logbuf_head = serial_logbuf;
+return 0;
+}
+
+void serial_mem_setbrg(void)
+{
+}
+
+int serial_mem_tstc(void)
+{
+return 0;
+}
+
+int serial_mem_getc(void)
+{
+return 0;
+}
+
+void serial_mem_putc(const char c)
+{
+*serial_logbuf_head = c;
+if (++serial_logbuf_head == serial_logbuf + CONFIG_UART_MEM)
+serial_logbuf_head = serial_lo

Re: [U-Boot] [PATCH v1 2/2] blackfin: bf609: addsoftswitch config command

Sonic Zhang — 2013-05-11T14:50:23

Hi Tom


On Wed, Feb 27, 2013 at 4:37 PM, Sonic Zhang <sonic.adi-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:

This is the patch that include "CONFIG_CMD_SOFTSWITCH". Is this what
you think that I forgot to post before?

Regards,

Sonic

Re: [U-Boot] Pull request: u-boot-blackfin

Tom Rini — 2013-05-09T12:09:38


Some number of these changes haven't been posted to the list for review
yet.  I'm pretty sure.  I don't recall seeing the CONFIG_CMD_SOFTSWITCH
stuff go past, and you didn't document that new flag in the top-level
README.  Also, you did a merge of your patman branch which means there's
extra stuff in your commit messages.

Please submit the new patches to the mailing list, and re-submit the
pull request with everything else you do have ready.  Thanks!

Pull request: u-boot-blackfin

Sonic Zhang — 2013-05-09T03:13:52

Hi Tom,

Please pull the following patches from u-boot-blackfin into your tree.

Thanks

Sonic

The following changes since commit fe8e4dbad1d31c452af3fbabba21e72b210381b3:

  nds32: Use sections header to obtain link symbols (2013-05-08 12:38:10 +0800)

are available in the git repository at:

  git://git.denx.de/u-boot-blackfin.git master

for you to fetch changes up to 2ac75146988bdcf8c30aee9c286af7f2f93b474e:

  bfin: Move gpio support for bf54x and bf60x into the generic driver
folder. (2013-05-09 10:53:47 +0800)

----------------------------------------------------------------
Bob Liu (3):
      blackfin: reduce size of u-boot.ldr in bf548-ezkit default config.
      blackfin: add baudrate to bdinfo
      blackfin: bf609: add softswitch config command

Marek Vasut (1):
      blackfin: The buf variable in bfin_mac.c is not used and produces warning,

Mike Frysinger (1):
      Blackfin: adjust asm constraints with NMI workaround

Scott Jiang (3):
      bfin: Remove spi dma function in bf5xx.
      bfin:

Re: [PATCH v1] bfin: Move gpio support for bf54x and bf60x into the generic driver folder.

Sonic Zhang — 2013-05-03T04:53:21

Hi Mike,

On Fri, May 3, 2013 at 12:23 AM, Mike Frysinger <vapier-aBrp7R+bbdUdnm+yROfE0A< at >public.gmane.org> wrote:

Don't worry. I will do it in Linux kernel soon after this is merged into u-boot.

Regards,

Sonic

Re: [PATCH v1] mmc: bfin: Ensure MMR writing is donebefore next command.

Sonic Zhang — 2013-04-27T11:21:47

Hi Mike

On Sat, Apr 27, 2013 at 7:14 PM, Mike Frysinger <vapier-aBrp7R+bbdUdnm+yROfE0A< at >public.gmane.org> wrote:

Good catch. I can't find the log of this patch in local tree, because
you moved the bfin_sdh.c someday.
Yes, this patch looks like a workaround. So, ignore it.

Regards,

Sonic

Re: [PATCH v1] mmc: bfin: Ensure MMR writing is donebefore next command.

Mike Frysinger — 2013-04-27T11:14:49

this patch was written by Cliff, not yourself.  the description really doesn't 
match reality, and no one has yet to explain what this is doing.  i'm fairly 
certain Cliff, when presented with the bug (iirc, it was "mmc doesn't work at 
low SCLK") just threw in a crap ton of SSYNCs to slow the driver down and 
shuffled some code until it worked.  no real analysis as to what was actually 
necessary.

hence, i never pushed this upstream.  i still don't think this belongs in 
mainline.
-mike
_______________________________________________
U-Boot-devel mailing list
U-Boot-devel-ZG0+EudsQA8dtHy/vicBwGD2FQJk+8+b< at >public.gmane.org
https://blackfin.uclinux.org/mailman/listinfo/u-boot-devel

[PATCH v1] mmc: bfin: Ensure MMR writing is donebefore next command.

Sonic Zhang — 2013-04-27T11:04:10

From: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>

- Ensure MMR writing is done before next command.
- Invalidate the buffer before starting to read.

Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
---
 drivers/mmc/bfin_sdh.c |   30 +++++++++++++++++++++++-------
 1 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/drivers/mmc/bfin_sdh.c b/drivers/mmc/bfin_sdh.c
index 0f98b96..028cb58 100644
--- a/drivers/mmc/bfin_sdh.c
+++ b/drivers/mmc/bfin_sdh.c
< at >< at > -136,6 +136,9 < at >< at > static int sdh_setup_data(struct mmc *mmc, struct mmc_data *data)
 /* Don't support write yet. */
 if (data->flags & MMC_DATA_WRITE)
 return UNUSABLE_ERR;
+blackfin_dcache_flush_invalidate_range(data->dest,
+data->dest + data_size);
+SSYNC();
 #ifndef RSI_BLKSZ
 data_ctl |= ((ffs(data_size) - 1) << 4);
 #else
< at >< at > -143,21 +146,22 < at >< at > static int sdh_setup_data(struct mmc *mmc, struct mmc_data *data)
 #endif
 data_ctl |= DTX_DIR;
 bfin_write_SDH_DATA_CTL(data_ctl);
+SSY

[PATCH v1] bf609: add SPI register base address

Sonic Zhang — 2013-04-01T09:02:58

From: Scott Jiang <scott.jiang.linux-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

- BF609 spi driver depend on this.

Signed-off-by: Scott Jiang <scott.jiang.linux-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>
Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
---
 arch/blackfin/include/asm/mach-bf609/BF609_def.h |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/arch/blackfin/include/asm/mach-bf609/BF609_def.h b/arch/blackfin/include/asm/mach-bf609/BF609_def.h
index 4d3b003..02b81d3 100644
--- a/arch/blackfin/include/asm/mach-bf609/BF609_def.h
+++ b/arch/blackfin/include/asm/mach-bf609/BF609_def.h
< at >< at > -128,6 +128,9 < at >< at >
 #define EMAC0_MACCFG      0xFFC20000 /* EMAC0 MAC Configuration Register */
 #define EMAC1_MACCFG      0xFFC22000 /* EMAC1 MAC Configuration Register */
 
+#define SPI0_REGBASE      0xFFC40400 /* SPI0 Base Address */
+#define SPI1_REGBASE      0xFFC40500 /* SPI1 Base Address */
+
 #define DMA10_DSCPTR_NXT  0xFFC05000 /* DMA10 Pointer to Next Initia

Re: [U-Boot] [PATCH v1] DOS_PBR block type is also valid dos block type.

Tom Rini — 2013-03-15T17:29:27

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 03/15/2013 01:09 PM, Stephen Warren wrote:

To me the question is, are we looking at a "torture card" that doesn't
really contain a compliant set of choices, but happens to not fail in
other cases (that don't play clever games like we do) or a really
valid card and we need to make a harder choice.

- -- 
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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Re: [U-Boot] [PATCH v1] DOS_PBR block type is also valid dos block type.

Tom Rini — 2013-03-15T13:14:49

Well, it sounds like this card has been put through a series of odd
formats and it's not strictly a valid card, but just not something that
doesn't break (because we're trying to be clever and it's causing us
problems).  Or do you have a number of cards which show this problem?

Re: [U-Boot] [PATCH v1] DOS_PBR block type is also valid dos block type.

Tom Rini — 2013-03-15T12:46:14

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 03/15/2013 01:10 AM, Stephen Warren wrote:

Thanks for looking into all of this.  What exactly fails with this
image, without this patch applied?

- -- 
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQIcBAEBAgAGBQJRQxgVAAoJENk4IS6UOR1WCtEQAIvLpSOD8ZgM2B0aKNUI92sr
ZgULRoQ7bjBL12NPrHQEdQn09cS3RTGY7AijWECfH29m5RN6e23tri5/MSRrw9yw
jxEPhIEWqNaJ0gEF9qteNm9QDlJyGq9AENMWIf33fqo+hF9jrAsFwzZKRmWnNm8U
Euc+KGdIwqBkD2ke+YIwHhV7ohII4LWrUD7FTYTYT78By3YR0YNpqrrgNuJB5Bs6
1TDARj/qTYk5uGy+1Ep8EqTIMqWfWtmeUhE5k6wYONOYPaETnDTezoxvN4wAuIyP
QF25XkJbNfU39UreGnXscWjeuS/3FHzEC6ArfT0n6uVv94fAbcH/w6gFUanNa3Ya
50/314K0ePiRH2rJP8/7tiCQixZtykvAfx5IfLr4HZsbDZtYMcLkKGvKiWkt+0Hp
MKKm059GK4W1tPbc31sFToqrGNNo41bE8hcDBa8rjFsbKkdDbXnMdMyRm4J5qWBX
MTGYsmw4veRVn8NAZVkCkyckZOIPu2I4CwLv4KyMBQZ/p/o70xzMMpOyBM6kXWWE
+8uoFZEL0jjJh214TUjRNu6YfzKyTiIriO6zKRsU0Do1iGGQ9EhiJxWKeLsuxz2i
C7o11B4ayFsrpJ+GfpV6JEl2UBRdBmiDmsp

[PATCH v1 1/3] blackfin: Enable early print via thegeneric serial API.

Sonic Zhang — 2013-03-14T07:29:20

From: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>

Remove blackfin specific implementation of the generic serial API when
early print macro is defined.

In BFIN_BOOT_BYPASS mode, don't call generic serial_puts, because
early print in bypass mode is running before code binary is relocated
to the link address.

Signed-off-by: Sonic Zhang <sonic.zhang-OyLXuOCK7orQT0dZR+AlfA< at >public.gmane.org>
---
 arch/blackfin/cpu/serial.c        |   71 ++++++++++++------------------------
 arch/blackfin/cpu/serial.h        |   10 +++---
 include/configs/bfin_adi_common.h |    3 --
 3 files changed, 29 insertions(+), 55 deletions(-)

diff --git a/arch/blackfin/cpu/serial.c b/arch/blackfin/cpu/serial.c
index 36d2a5c..89290d6 100644
--- a/arch/blackfin/cpu/serial.c
+++ b/arch/blackfin/cpu/serial.c
< at >< at > -197,13 +197,16 < at >< at > static void uart_loop(uint32_t uart_base, int state)
 
 static inline void __serial_set_baud(uint32_t uart_base, uint32_t baud)
 {
+#ifdef CONFIG_DEBUG_EARLY_SERIAL
+serial_early_set_baud(uar