[U-Boot] [PATCH v2] ppc4xx: Add Sequoia RAM-booting target

[U-Boot] [PATCH v2] ppc4xx: Add Sequoia RAM-booting target

[Stefan Roese]

This patch adds another build target for the AMCC Sequoia PPC440EPx
eval board. This RAM-booting version is targeted for boards without
NOR FLASH (NAND booting) which need a possibility to initially
program their NAND FLASH. Using a JTAG debugger (e.g. BDI2000/3000)
configured to setup the SDRAM, this debugger can load this RAM-
booting image to the target address in SDRAM (in this case 0x1000000)
and start it there. Then U-Boot's standard NAND commands can be
used to program the NAND FLASH (e.g. "nand write ...").

Here the commands to load and start this image from the BDI2000:

440EPX>load 0x1000000 /tftpboot/sequoia/u-boot.bin
440EPX>go 0x1000000

Please note that this image automatically scans for an already
initialized SDRAM TLB (detected by EPN=0). This TLB will not be
cleared. So your debugger should configure the SDRAM TLB correctly
(as done in the standard Sequoia BDI init script).

Signed-off-by: Stefan Roese <sr@denx.de>
---
Changes in v2:
- Using #ifdef's instead of __attribute__((unused)) for variable
  declatration.

 Makefile                          |   11 +++
 board/amcc/sequoia/init.S         |    7 ++
 board/amcc/sequoia/sdram.c        |    3 +-
 board/amcc/sequoia/sequoia.c      |   10 +++-
 board/amcc/sequoia/u-boot-ram.lds |  126 +++++++++++++++++++++++++++++++++++++
 cpu/ppc4xx/start.S                |   33 ++++++++--
 include/configs/amcc-common.h     |   11 +++
 include/configs/sequoia.h         |   30 +++++++--
 8 files changed, 215 insertions(+), 16 deletions(-)
 create mode 100644 board/amcc/sequoia/u-boot-ram.lds

diff --git a/Makefile b/Makefile
index 137c88f..1ae93ab 100644
--- a/Makefile
+++ b/Makefile
@@ -1533,6 +1533,17 @@ rainier_nand_config: unconfig
 	@echo "TEXT_BASE = 0x01000000" > $(obj)board/amcc/sequoia/config.tmp
 	@echo "CONFIG_NAND_U_BOOT = y" >> $(obj)include/config.mk
 
+sequoia_ramboot_config \
+rainier_ramboot_config: unconfig
+	@mkdir -p $(obj)include $(obj)board/amcc/sequoia
+	@echo "#define CONFIG_SYS_RAMBOOT" > $(obj)include/config.h
+	@echo "#define CONFIG_$$(echo $(subst ,,$(@:_config=)) | \
+		tr '[:lower:]' '[:upper:]')" >> $(obj)include/config.h
+	@$(MKCONFIG) -n $@ -a sequoia ppc ppc4xx sequoia amcc
+	@echo "TEXT_BASE = 0x01000000" > $(obj)board/amcc/sequoia/config.tmp
+	@echo "LDSCRIPT = board/amcc/sequoia/u-boot-ram.lds" >> \
+		$(obj)board/amcc/sequoia/config.tmp
+
 taihu_config:	unconfig
 	@$(MKCONFIG) $(@:_config=) ppc ppc4xx taihu amcc
 
diff --git a/board/amcc/sequoia/init.S b/board/amcc/sequoia/init.S
index 4452d26..3c0e400 100644
--- a/board/amcc/sequoia/init.S
+++ b/board/amcc/sequoia/init.S
@@ -43,12 +43,19 @@ tlbtab:
 	/* vxWorks needs this as first entry for the Machine Check interrupt */
 	tlbentry( 0x40000000, SZ_256M, 0, 0, AC_R|AC_W|AC_X|SA_G|SA_I )
 
+	/*
+	 * The RAM-boot version skips the SDRAM TLB (identified by EPN=0). This
+	 * entry is already configured for SDRAM via the JTAG debugger and mustn't
+	 * be re-initialized by this RAM-booting U-Boot version.
+	 */
+#ifndef CONFIG_SYS_RAMBOOT
 	/* TLB-entry for DDR SDRAM (Up to 2GB) */
 #ifdef CONFIG_4xx_DCACHE
 	tlbentry( CONFIG_SYS_SDRAM_BASE, SZ_256M, CONFIG_SYS_SDRAM_BASE, 0, AC_R|AC_W|AC_X|SA_G)
 #else
 	tlbentry( CONFIG_SYS_SDRAM_BASE, SZ_256M, CONFIG_SYS_SDRAM_BASE, 0, AC_R|AC_W|AC_X|SA_G|SA_I )
 #endif
+#endif /* CONFIG_SYS_RAMBOOT */
 
 	/* TLB-entry for EBC */
 	tlbentry( CONFIG_SYS_BCSR_BASE, SZ_256M, CONFIG_SYS_BCSR_BASE, 1, AC_R|AC_W|AC_X|SA_G|SA_I )
diff --git a/board/amcc/sequoia/sdram.c b/board/amcc/sequoia/sdram.c
index 6df4c6d..bde471c 100644
--- a/board/amcc/sequoia/sdram.c
+++ b/board/amcc/sequoia/sdram.c
@@ -54,7 +54,8 @@ extern void denali_core_search_data_eye(void);
  ************************************************************************/
 phys_size_t initdram (int board_type)
 {
-#if !defined(CONFIG_NAND_U_BOOT) || defined(CONFIG_NAND_SPL)
+#if !(defined(CONFIG_NAND_U_BOOT) || defined(CONFIG_SYS_RAMBOOT)) || \
+    defined(CONFIG_NAND_SPL)
 	ulong speed = get_bus_freq(0);
 
 	mtsdram(DDR0_02, 0x00000000);
diff --git a/board/amcc/sequoia/sequoia.c b/board/amcc/sequoia/sequoia.c
index e824b8f..246ad94 100644
--- a/board/amcc/sequoia/sequoia.c
+++ b/board/amcc/sequoia/sequoia.c
@@ -33,7 +33,9 @@
 
 DECLARE_GLOBAL_DATA_PTR;
 
+#if !defined(CONFIG_SYS_NO_FLASH)
 extern flash_info_t flash_info[CONFIG_SYS_MAX_FLASH_BANKS]; /* info for FLASH chips */
+#endif
 
 extern void __ft_board_setup(void *blob, bd_t *bd);
 ulong flash_get_size(ulong base, int banknum);
@@ -122,16 +124,19 @@ int board_early_init_f(void)
 
 int misc_init_r(void)
 {
+#if !defined(CONFIG_SYS_NO_FLASH)
 	uint pbcr;
 	int size_val = 0;
-	u32 reg;
+#endif
 #ifdef CONFIG_440EPX
 	unsigned long usb2d0cr = 0;
 	unsigned long usb2phy0cr, usb2h0cr = 0;
 	unsigned long sdr0_pfc1;
 	char *act = getenv("usbact");
 #endif
+	u32 reg;
 
+#if !defined(CONFIG_SYS_NO_FLASH)
 	/* Re-do flash sizing to get full correct info */
 
 	/* adjust flash start and offset */
@@ -171,6 +176,7 @@ int misc_init_r(void)
 			    CONFIG_ENV_ADDR_REDUND + 2*CONFIG_ENV_SECT_SIZE - 1,
 			    &flash_info[0]);
 #endif
+#endif /* CONFIG_SYS_NO_FLASH */
 
 	/*
 	 * USB suff...
@@ -515,7 +521,7 @@ int post_hotkeys_pressed(void)
 }
 #endif /* CONFIG_POST */
 
-#if defined(CONFIG_NAND_U_BOOT)
+#if defined(CONFIG_NAND_U_BOOT) || defined(CONFIG_SYS_RAMBOOT)
 /*
  * On NAND-booting sequoia, we need to patch the chips select numbers
  * in the dtb (CS0 - NAND, CS3 - NOR)
diff --git a/board/amcc/sequoia/u-boot-ram.lds b/board/amcc/sequoia/u-boot-ram.lds
new file mode 100644
index 0000000..9393b65
--- /dev/null
+++ b/board/amcc/sequoia/u-boot-ram.lds
@@ -0,0 +1,126 @@
+/*
+ * (C) Copyright 2009
+ * Stefan Roese, DENX Software Engineering, sr at denx.de.
+ *
+ * See file CREDITS for list of people who contributed to this
+ * project.
+ *
+ * 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.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+OUTPUT_ARCH(powerpc)
+SECTIONS
+{
+  /* Read-only sections, merged into text segment: */
+  . = + SIZEOF_HEADERS;
+  .interp : { *(.interp) }
+  .hash          : { *(.hash)		}
+  .dynsym        : { *(.dynsym)		}
+  .dynstr        : { *(.dynstr)		}
+  .rel.text      : { *(.rel.text)		}
+  .rela.text     : { *(.rela.text)	}
+  .rel.data      : { *(.rel.data)		}
+  .rela.data     : { *(.rela.data)	}
+  .rel.rodata    : { *(.rel.rodata)	}
+  .rela.rodata   : { *(.rela.rodata)	}
+  .rel.got       : { *(.rel.got)		}
+  .rela.got      : { *(.rela.got)		}
+  .rel.ctors     : { *(.rel.ctors)	}
+  .rela.ctors    : { *(.rela.ctors)	}
+  .rel.dtors     : { *(.rel.dtors)	}
+  .rela.dtors    : { *(.rela.dtors)	}
+  .rel.bss       : { *(.rel.bss)		}
+  .rela.bss      : { *(.rela.bss)		}
+  .rel.plt       : { *(.rel.plt)		}
+  .rela.plt      : { *(.rela.plt)		}
+  .init          : { *(.init)	}
+  .plt : { *(.plt) }
+  .text      :
+  {
+    cpu/ppc4xx/start.o	(.text)
+
+    *(.text)
+    *(.fixup)
+    *(.got1)
+  }
+  _etext = .;
+  PROVIDE (etext = .);
+  .rodata    :
+  {
+    *(SORT_BY_ALIGNMENT(SORT_BY_NAME(.rodata*)))
+  }
+  .fini      : { *(.fini)    } =0
+  .ctors     : { *(.ctors)   }
+  .dtors     : { *(.dtors)   }
+
+  /* Read-write section, merged into data segment: */
+  . = (. + 0x00FF) & 0xFFFFFF00;
+  _erotext = .;
+  PROVIDE (erotext = .);
+  .reloc   :
+  {
+    *(.got)
+    _GOT2_TABLE_ = .;
+    *(.got2)
+    _FIXUP_TABLE_ = .;
+    *(.fixup)
+  }
+  __got2_entries = (_FIXUP_TABLE_ - _GOT2_TABLE_) >>2;
+  __fixup_entries = (. - _FIXUP_TABLE_)>>2;
+
+  .data    :
+  {
+    *(.data)
+    *(.data1)
+    *(.sdata)
+    *(.sdata2)
+    *(.dynamic)
+    CONSTRUCTORS
+  }
+  _edata  =  .;
+  PROVIDE (edata = .);
+
+  . = .;
+  __u_boot_cmd_start = .;
+  .u_boot_cmd : { *(.u_boot_cmd) }
+  __u_boot_cmd_end = .;
+
+
+  . = .;
+  __start___ex_table = .;
+  __ex_table : { *(__ex_table) }
+  __stop___ex_table = .;
+
+  . = ALIGN(256);
+  __init_begin = .;
+  .text.init : { *(.text.init) }
+  .data.init : { *(.data.init) }
+  . = ALIGN(256);
+  __init_end = .;
+
+  __bss_start = .;
+  .bss (NOLOAD)       :
+  {
+   *(.sbss) *(.scommon)
+   *(.dynbss)
+   *(.bss)
+   *(COMMON)
+   . = ALIGN(4);
+  }
+
+  _end = . ;
+  PROVIDE (end = .);
+}
diff --git a/cpu/ppc4xx/start.S b/cpu/ppc4xx/start.S
index f2b8908..ac96fc2 100644
--- a/cpu/ppc4xx/start.S
+++ b/cpu/ppc4xx/start.S
@@ -257,6 +257,14 @@
 	bl	board_init_f
 #endif
 
+#if defined(CONFIG_SYS_RAMBOOT)
+	/*
+	 * 4xx RAM-booting U-Boot image is started from offset 0
+	 */
+	.text
+	bl	_start_440
+#endif
+
 /*
  * 440 Startup -- on reset only the top 4k of the effective
  * address space is mapped in by an entry in the instruction
@@ -444,10 +452,17 @@ skip_debug_init:
 	addis	r0,0,0x0000
 	li	r1,0x003f	/* 64 TLB entries */
 	mtctr	r1
-rsttlb:	tlbwe	r0,r1,0x0000	/* Invalidate all entries (V=0)*/
-	tlbwe	r0,r1,0x0001
-	tlbwe	r0,r1,0x0002
-	subi	r1,r1,0x0001
+	li	r4,0		/* Start with TLB #0 */
+rsttlb:
+#ifdef CONFIG_SYS_RAMBOOT
+	tlbre	r3,r4,0		/* Read contents from TLB word #0 to get EPN */
+	rlwinm.	r3,r3,0,0xfffffc00	/* Mask EPN */
+	beq	tlbnxt		/* Skip EPN=0 TLB, this is the SDRAM TLB */
+#endif
+	tlbwe	r0,r4,0		/* Invalidate all entries (V=0)*/
+	tlbwe	r0,r4,1
+	tlbwe	r0,r4,2
+tlbnxt:	addi	r4,r4,1		/* Next TLB */
 	bdnz	rsttlb
 
 	/*----------------------------------------------------------------*/
@@ -476,7 +491,13 @@ rsttlb:	tlbwe	r0,r1,0x0000	/* Invalidate all entries (V=0)*/
 	li	r4,0		/* TLB # */
 
 	addi	r5,r5,-4
-1:	lwzu	r0,4(r5)
+1:
+#ifdef CONFIG_SYS_RAMBOOT
+	tlbre	r3,r4,0		/* Read contents from TLB word #0 */
+	rlwinm.	r3,r3,0,0x00000200	/* Mask V (valid) bit */
+	bne	tlbnx2		/* Skip V=1 TLB, this is the SDRAM TLB */
+#endif
+	lwzu	r0,4(r5)
 	cmpwi	r0,0
 	beq	2f		/* 0 marks end */
 	lwzu	r1,4(r5)
@@ -484,7 +505,7 @@ rsttlb:	tlbwe	r0,r1,0x0000	/* Invalidate all entries (V=0)*/
 	tlbwe	r0,r4,0		/* TLB Word 0 */
 	tlbwe	r1,r4,1		/* TLB Word 1 */
 	tlbwe	r2,r4,2		/* TLB Word 2 */
-	addi	r4,r4,1		/* Next TLB */
+tlbnx2:	addi	r4,r4,1		/* Next TLB */
 	bdnz	1b
 
 	/*----------------------------------------------------------------*/
diff --git a/include/configs/amcc-common.h b/include/configs/amcc-common.h
index d3dc3e5..3b733c0 100644
--- a/include/configs/amcc-common.h
+++ b/include/configs/amcc-common.h
@@ -76,6 +76,17 @@
 #define CONFIG_CMD_PING
 #define CONFIG_CMD_REGINFO
 
+#if defined(CONFIG_SYS_RAMBOOT)
+/*
+ * Disable NOR FLASH commands on RAM-booting version. One main reason for this
+ * RAM-booting version is boards with NAND and without NOR. This image can
+ * be used for initial NAND programming.
+ */
+#define CONFIG_SYS_NO_FLASH
+#undef CONFIG_CMD_FLASH
+#undef CONFIG_CMD_IMLS
+#endif
+
 /*
  * Miscellaneous configurable options
  */
diff --git a/include/configs/sequoia.h b/include/configs/sequoia.h
index fa226b2..89acacc 100644
--- a/include/configs/sequoia.h
+++ b/include/configs/sequoia.h
@@ -112,13 +112,26 @@
 /*
  * Environment
  */
-#if !defined(CONFIG_NAND_U_BOOT) && !defined(CONFIG_NAND_SPL)
-#define CONFIG_ENV_IS_IN_FLASH	1	/* use FLASH for environ vars	*/
+#if defined(CONFIG_NAND_U_BOOT) || defined(CONFIG_NAND_SPL)
+#define CONFIG_ENV_IS_IN_NAND		/* use NAND for environ vars	*/
+#define CONFIG_ENV_IS_EMBEDDED		/* use embedded environment	*/
+#elif defined(CONFIG_SYS_RAMBOOT)
+#define CONFIG_ENV_IS_NOWHERE		/* Store env in memory only	*/
+#define CONFIG_ENV_SIZE		(8 << 10)
+/*
+ * In RAM-booting version, we have no environment storage. So we need to
+ * provide@least preliminary MAC addresses for the 4xx EMAC driver to
+ * register the interfaces. Those two addresses are generated via the
+ * tools/gen_eth_addr tool and should only be used in a closed laboratory
+ * environment.
+ */
+#define	CONFIG_ETHADDR		4a:56:49:22:3e:43
+#define	CONFIG_ETH1ADDR		02:93:53:d5:06:98
 #else
-#define CONFIG_ENV_IS_IN_NAND	1	/* use NAND for environ vars	*/
-#define CONFIG_ENV_IS_EMBEDDED	1	/* use embedded environment	*/
+#define CONFIG_ENV_IS_IN_FLASH		/* use FLASH for environ vars	*/
 #endif
 
+#if defined(CONFIG_CMD_FLASH)
 /*
  * FLASH related
  */
@@ -148,6 +161,7 @@
 #define CONFIG_ENV_ADDR_REDUND	(CONFIG_ENV_ADDR-CONFIG_ENV_SECT_SIZE)
 #define CONFIG_ENV_SIZE_REDUND	(CONFIG_ENV_SIZE)
 #endif
+#endif /* CONFIG_CMD_FLASH */
 
 /*
  * IPL (Initial Program Loader, integrated inside CPU)
@@ -211,7 +225,8 @@
  * DDR SDRAM
  */
 #define CONFIG_SYS_MBYTES_SDRAM        (256)	/* 256MB			*/
-#if !defined(CONFIG_NAND_U_BOOT) && !defined(CONFIG_NAND_SPL)
+#if !defined(CONFIG_NAND_U_BOOT) && !defined(CONFIG_NAND_SPL) && \
+    !defined(CONFIG_SYS_RAMBOOT)
 #define CONFIG_DDR_DATA_EYE		/* use DDR2 optimization	*/
 #endif
 #define CONFIG_SYS_MEM_TOP_HIDE	(4 << 10) /* don't use last 4kbytes	*/
@@ -306,7 +321,7 @@
  * overwrite part of the U-Boot image which is already loaded from NAND
  * to SDRAM.
  */
-#if defined(CONFIG_NAND_U_BOOT)
+#if defined(CONFIG_NAND_U_BOOT) || defined(CONFIG_SYS_RAMBOOT)
 #define CONFIG_SYS_POST_MEMORY_ON	0
 #else
 #define CONFIG_SYS_POST_MEMORY_ON	CONFIG_SYS_POST_MEMORY
@@ -354,7 +369,8 @@
 /*
  * On Sequoia CS0 and CS3 are switched when configuring for NAND booting
  */
-#if !defined(CONFIG_NAND_U_BOOT) && !defined(CONFIG_NAND_SPL)
+#if !defined(CONFIG_NAND_U_BOOT) && !defined(CONFIG_NAND_SPL) && \
+    !defined(CONFIG_SYS_RAMBOOT)
 #define CONFIG_SYS_NAND_CS		3	/* NAND chip connected to CSx	*/
 /* Memory Bank 0 (NOR-FLASH) initialization				*/
 #define CONFIG_SYS_EBC_PB0AP		0x03017200
-- 
1.6.2.5

[U-Boot] [PATCH v2] ppc4xx: Add Sequoia RAM-booting target

[Wolfgang Denk]

Dear Stefan Roese,

In message <1242035235-26833-1-git-send-email-sr@denx.de> you wrote:
>
> Here the commands to load and start this image from the BDI2000:
> 
> 440EPX>load 0x1000000 /tftpboot/sequoia/u-boot.bin
> 440EPX>go 0x1000000

I assume you used "reset halt" before?

> Please note that this image automatically scans for an already
> initialized SDRAM TLB (detected by EPN=0). This TLB will not be

And what happens when the TLB is not correctly set? Will we get a
useful error message then?

> cleared. So your debugger should configure the SDRAM TLB correctly
> (as done in the standard Sequoia BDI init script).

Well, standard procedure is to boot using "reset run"; I think you
should add the "reset halt" step to the documentation above - just to
be sure.

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
"There is no statute of limitations on stupidity."
- Randomly produced by a computer program called Markov3.

[U-Boot] [PATCH v2] ppc4xx: Add Sequoia RAM-booting target

[Stefan Roese]

Hi Wolfgang,

On Monday 11 May 2009 12:21:23 Wolfgang Denk wrote:
> > Here the commands to load and start this image from the BDI2000:
> >
> > 440EPX>load 0x1000000 /tftpboot/sequoia/u-boot.bin
> > 440EPX>go 0x1000000
>
> I assume you used "reset halt" before?

Yes, sure.

> > Please note that this image automatically scans for an already
> > initialized SDRAM TLB (detected by EPN=0). This TLB will not be
>
> And what happens when the TLB is not correctly set? Will we get a
> useful error message then?

No. SDRAM needs to work correctly so that this RAM-booting version can 
function. And for this the MMU needs to be configured and the SDRAM controller 
setup correctly. If this is not the case and SDRAM is not working from the 
debugger then we have no chance to load this image without problems (verify 
will most likely fail) or run it.

> > cleared. So your debugger should configure the SDRAM TLB correctly
> > (as done in the standard Sequoia BDI init script).
>
> Well, standard procedure is to boot using "reset run"; I think you
> should add the "reset halt" step to the documentation above - just to
> be sure.

Yes good idea. I'll update the commit text and resubmit.

Best regards,
Stefan

=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80  Email: office at denx.de
=====================================================================

[U-Boot] [PATCH v2] ppc4xx: Add Sequoia RAM-booting target

[Wolfgang Denk]

Dear Stefan Roese,

In message <200905111339.28531.sr@denx.de> you wrote:
> 
> > > Please note that this image automatically scans for an already
> > > initialized SDRAM TLB (detected by EPN=0). This TLB will not be
> >
> > And what happens when the TLB is not correctly set? Will we get a
> > useful error message then?
> 
> No. SDRAM needs to work correctly so that this RAM-booting version can 
> function. And for this the MMU needs to be configured and the SDRAM controller 
> setup correctly. If this is not the case and SDRAM is not working from the 
> debugger then we have no chance to load this image without problems (verify 
> will most likely fail) or run it.

Hm... if the image cannot run anyway without initialized SDRAM TLB,
then why do you scan for it?  My understanding is that the result
wouldbe the same even if you omitted this test: without initialized
SDRAM TLB the system would just hang / crash without error message.
So this test is dead code.

Or am I missing something?

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Our universe is a fragile house of atoms, held together by the mortar
of cause-and-effect. One magician would be two too many.
                        - Terry Pratchett, _The Dark Side of the Sun_

[U-Boot] [PATCH v2] ppc4xx: Add Sequoia RAM-booting target

[Stefan Roese]

On Monday 11 May 2009 15:20:10 Wolfgang Denk wrote:
> > > And what happens when the TLB is not correctly set? Will we get a
> > > useful error message then?
> >
> > No. SDRAM needs to work correctly so that this RAM-booting version can
> > function. And for this the MMU needs to be configured and the SDRAM
> > controller setup correctly. If this is not the case and SDRAM is not
> > working from the debugger then we have no chance to load this image
> > without problems (verify will most likely fail) or run it.
>
> Hm... if the image cannot run anyway without initialized SDRAM TLB,
> then why do you scan for it?  My understanding is that the result
> wouldbe the same even if you omitted this test: without initialized
> SDRAM TLB the system would just hang / crash without error message.
> So this test is dead code.
>
> Or am I missing something?

I'm "scanning" for the SDRAM TLB, because it can be on any TLB location. In my 
first implementation I assumed that the SDRAM TLB was TLB #0 (configured via 
BDI2000). This worked just fine. But then I loaded this RAM-booting image via 
U-Boot and started it here (via go ...). And it crashed since U-Boot uses a 
different TLB #. So now I'm scanning/detecting the SDRAM TLB to be independent 
from the originator of the TLB. 

Best regards,
Stefan

=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80  Email: office at denx.de
=====================================================================

[U-Boot] [PATCH v2] ppc4xx: Add Sequoia RAM-booting target

[Wolfgang Denk]

Dear Stefan Roese,

In message <200905111531.11474.sr@denx.de> you wrote:
> On Monday 11 May 2009 15:20:10 Wolfgang Denk wrote:
> > > > And what happens when the TLB is not correctly set? Will we get a
> > > > useful error message then?
> > >
> > > No. SDRAM needs to work correctly so that this RAM-booting version can
> > > function. And for this the MMU needs to be configured and the SDRAM
> > > controller setup correctly. If this is not the case and SDRAM is not
> > > working from the debugger then we have no chance to load this image
> > > without problems (verify will most likely fail) or run it.
> >
> > Hm... if the image cannot run anyway without initialized SDRAM TLB,
> > then why do you scan for it?  My understanding is that the result
> > wouldbe the same even if you omitted this test: without initialized
> > SDRAM TLB the system would just hang / crash without error message.
> > So this test is dead code.
> >
> > Or am I missing something?
> 
> I'm "scanning" for the SDRAM TLB, because it can be on any TLB location. In my 
> first implementation I assumed that the SDRAM TLB was TLB #0 (configured via 
> BDI2000). This worked just fine. But then I loaded this RAM-booting image via 
> U-Boot and started it here (via go ...). And it crashed since U-Boot uses a 
> different TLB #. So now I'm scanning/detecting the SDRAM TLB to be independent 
> from the originator of the TLB. 

Ah, I see. Could you please put this information in the commit
message, too? 

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
: 1.  What is the possibility of this being added in the future?
In the near future, the probability is close to zero. In the  distant
future, I'll be dead, and posterity can do whatever they like... :-)
                                                              - lwall

[U-Boot] [PATCH v2] ppc4xx: Add Sequoia RAM-booting target

[Stefan Roese]

On Monday 11 May 2009 17:57:47 Wolfgang Denk wrote:
> > I'm "scanning" for the SDRAM TLB, because it can be on any TLB location.
> > In my first implementation I assumed that the SDRAM TLB was TLB #0
> > (configured via BDI2000). This worked just fine. But then I loaded this
> > RAM-booting image via U-Boot and started it here (via go ...). And it
> > crashed since U-Boot uses a different TLB #. So now I'm
> > scanning/detecting the SDRAM TLB to be independent from the originator of
> > the TLB.
>
> Ah, I see. Could you please put this information in the commit
> message, too?

I thought it was obvious. :) OK, I'll add it to the commit text.

Best regards,
Stefan

=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80  Email: office at denx.de
=====================================================================