[PATCH 00/11] ppc32 mm init clean and 85xx kernel reloc

[PATCH 00/11] ppc32 mm init clean and 85xx kernel reloc

[Kumar Gala]

This set of patches cleans up the initialization of various mmu bits on
ppc32 (and a small bit on ppc64 to maintain common code) towards the
goal of having an 85xx (book-e) kernel able to run at non-zero offsets.

These patches exist in my master and ppc32_mm_init branch.  I've dropped
them from the powerpc-next as paulus wants them to go via his tree
directly since they touch common code.

- k

[PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Kumar Gala]

Normally we assume kernel images will be loaded at offset 0. However
there are situations, like when the kernel itself is running at a non-zero
physical address, that we don't want to load it at 0.

Allow the wrapper to take an offset.  We use this when building u-boot images.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/boot/Makefile |    7 +++++++
 arch/powerpc/boot/wrapper  |   12 ++++++++++--
 2 files changed, 17 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile
index a5528ab..3c80858 100644
--- a/arch/powerpc/boot/Makefile
+++ b/arch/powerpc/boot/Makefile
@@ -35,6 +35,12 @@ endif
 
 BOOTCFLAGS	+= -I$(obj) -I$(srctree)/$(obj) -I$(srctree)/$(src)/libfdt
 
+ifdef CONFIG_MEMORY_START
+MEMBASE=$(CONFIG_MEMORY_START)
+else
+MEMBASE=0x00000000
+endif
+
 $(obj)/4xx.o: BOOTCFLAGS += -mcpu=405
 $(obj)/ebony.o: BOOTCFLAGS += -mcpu=405
 $(obj)/cuboot-taishan.o: BOOTCFLAGS += -mcpu=405
@@ -181,6 +187,7 @@ endif
 # args (to if_changed): 1 = (this rule), 2 = platform, 3 = dts 4=dtb 5=initrd
 quiet_cmd_wrap	= WRAP    $@
       cmd_wrap	=$(CONFIG_SHELL) $(wrapper) -c -o $@ -p $2 $(CROSSWRAP) \
+		-m $(MEMBASE) \
 		$(if $3, -s $3)$(if $4, -d $4)$(if $5, -i $5) vmlinux
 
 image-$(CONFIG_PPC_PSERIES)		+= zImage.pseries
diff --git a/arch/powerpc/boot/wrapper b/arch/powerpc/boot/wrapper
index 03b474b..011d148 100755
--- a/arch/powerpc/boot/wrapper
+++ b/arch/powerpc/boot/wrapper
@@ -50,8 +50,11 @@ objbin=$object
 # directory for working files
 tmpdir=.
 
+# physical offset of kernel image
+membase=0x00000000
+
 usage() {
-    echo 'Usage: wrapper [-o output] [-p platform] [-i initrd]' >&2
+    echo 'Usage: wrapper [-o output] [-p platform] [-i initrd] [-m membase]' >&2
     echo '       [-d devtree] [-s tree.dts] [-c] [-C cross-prefix]' >&2
     echo '       [-D datadir] [-W workingdir] [--no-gzip] [vmlinux]' >&2
     exit 1
@@ -84,6 +87,11 @@ while [ "$#" -gt 0 ]; do
 	[ "$#" -gt 0 ] || usage
 	dts="$1"
 	;;
+    -m)
+	shift
+	[ "$#" -gt 0 ] || usage
+	membase="$1"
+	;;
     -c)
 	cacheit=y
 	;;
@@ -229,7 +237,7 @@ fi
 case "$platform" in
 uboot)
     rm -f "$ofile"
-    mkimage -A ppc -O linux -T kernel -C gzip -a 00000000 -e 00000000 \
+    mkimage -A ppc -O linux -T kernel -C gzip -a $membase -e $membase \
 	$uboot_version -d "$vmz" "$ofile"
     if [ -z "$cacheit" ]; then
 	rm -f "$vmz"
-- 
1.5.4.1

[PATCH 02/11] [POWERPC] Remove Kconfig option BOOT_LOAD

[Kumar Gala]

Nothing appears to use BOOT_LOAD so remove it as a configurable option.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/Kconfig |   16 ----------------
 1 files changed, 0 insertions(+), 16 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 3651355..4d9ced2 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -674,22 +674,6 @@ config CONSISTENT_SIZE
 	hex "Size of consistent memory pool" if CONSISTENT_SIZE_BOOL
 	default "0x00200000" if NOT_COHERENT_CACHE
 
-config BOOT_LOAD_BOOL
-	bool "Set the boot link/load address"
-	depends on ADVANCED_OPTIONS && !PPC_MULTIPLATFORM
-	help
-	  This option allows you to set the initial load address of the zImage
-	  or zImage.initrd file.  This can be useful if you are on a board
-	  which has a small amount of memory.
-
-	  Say N here unless you know what you are doing.
-
-config BOOT_LOAD
-	hex "Link/load address for booting" if BOOT_LOAD_BOOL
-	default "0x00400000" if 40x || 8xx || 8260
-	default "0x01000000" if 44x
-	default "0x00800000"
-
 config PIN_TLB
 	bool "Pinned Kernel TLBs (860 ONLY)"
 	depends on ADVANCED_OPTIONS && 8xx
-- 
1.5.4.1

[PATCH 03/11] [POWERPC] Provide access to arch/powerpc include path on ppc64

[Kumar Gala]

There does not appear to be any reason that we shouldn't just have
-Iarch/$(ARCH) on both ppc32 and ppc64 builds.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/Makefile |   10 ++++------
 1 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile
index ab5cfe8..43ee815 100644
--- a/arch/powerpc/Makefile
+++ b/arch/powerpc/Makefile
@@ -71,13 +71,11 @@ endif
 
 LDFLAGS_vmlinux	:= -Bstatic
 
-CPPFLAGS-$(CONFIG_PPC32) := -Iarch/$(ARCH)
-AFLAGS-$(CONFIG_PPC32)	:= -Iarch/$(ARCH)
 CFLAGS-$(CONFIG_PPC64)	:= -mminimal-toc -mtraceback=none  -mcall-aixdesc
-CFLAGS-$(CONFIG_PPC32)	:= -Iarch/$(ARCH) -ffixed-r2 -mmultiple
-KBUILD_CPPFLAGS	+= $(CPPFLAGS-y)
-KBUILD_AFLAGS	+= $(AFLAGS-y)
-KBUILD_CFLAGS	+= -msoft-float -pipe $(CFLAGS-y)
+CFLAGS-$(CONFIG_PPC32)	:= -ffixed-r2 -mmultiple
+KBUILD_CPPFLAGS	+= -Iarch/$(ARCH)
+KBUILD_AFLAGS	+= -Iarch/$(ARCH)
+KBUILD_CFLAGS	+= -msoft-float -pipe -Iarch/$(ARCH) $(CFLAGS-y)
 CPP		= $(CC) -E $(KBUILD_CFLAGS)
 
 CHECKFLAGS	+= -m$(CONFIG_WORD_SIZE) -D__powerpc__ -D__powerpc$(CONFIG_WORD_SIZE)__
-- 
1.5.4.1

[PATCH 04/11 v2] [POWERPC] Remove and replace uses of PPC_MEMSTART with memstart_addr

[Kumar Gala]

A number of users of PPC_MEMSTART (40x, ppc_mmu_32) can just always use
0 as we don't support booting these kernels at non-zero physical addresses
since their exception vectors must be at 0 (or 0xfffx_xxxx).

For the sub-arches that support relocatable interrupt vectors (book-e) its
reasonable to have memory start at a non-zero physical address.  For those
cases use the variable memstart_addr instead of the #define PPC_MEMSTART
since the only uses of PPC_MEMSTART are for initialization and in the
future we can set memstart_addr at runtime to have a relocatable kernel.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---

Fixed warning in ppc_mmu_32.c due to align still being defined.

 arch/powerpc/mm/40x_mmu.c       |    2 +-
 arch/powerpc/mm/fsl_booke_mmu.c |   11 +++++------
 arch/powerpc/mm/init_32.c       |    8 ++++----
 arch/powerpc/mm/mmu_decl.h      |    1 +
 arch/powerpc/mm/pgtable_32.c    |    5 +++--
 arch/powerpc/mm/ppc_mmu_32.c    |   11 ++---------
 include/asm-powerpc/page_32.h   |    2 --
 7 files changed, 16 insertions(+), 24 deletions(-)

diff --git a/arch/powerpc/mm/40x_mmu.c b/arch/powerpc/mm/40x_mmu.c
index 3899ea9..cecbbc7 100644
--- a/arch/powerpc/mm/40x_mmu.c
+++ b/arch/powerpc/mm/40x_mmu.c
@@ -97,7 +97,7 @@ unsigned long __init mmu_mapin_ram(void)
 	phys_addr_t p;
 
 	v = KERNELBASE;
-	p = PPC_MEMSTART;
+	p = 0;
 	s = total_lowmem;
 
 	if (__map_without_ltlbs)
diff --git a/arch/powerpc/mm/fsl_booke_mmu.c b/arch/powerpc/mm/fsl_booke_mmu.c
index c93a966..3dd0c81 100644
--- a/arch/powerpc/mm/fsl_booke_mmu.c
+++ b/arch/powerpc/mm/fsl_booke_mmu.c
@@ -53,13 +53,12 @@
 #include <asm/machdep.h>
 #include <asm/setup.h>
 
+#include "mmu_decl.h"
+
 extern void loadcam_entry(unsigned int index);
 unsigned int tlbcam_index;
 unsigned int num_tlbcam_entries;
 static unsigned long __cam0, __cam1, __cam2;
-extern unsigned long total_lowmem;
-extern unsigned long __max_low_memory;
-extern unsigned long __initial_memory_limit;
 #define MAX_LOW_MEM	CONFIG_LOWMEM_SIZE
 
 #define NUM_TLBCAMS	(16)
@@ -165,15 +164,15 @@ void invalidate_tlbcam_entry(int index)
 void __init cam_mapin_ram(unsigned long cam0, unsigned long cam1,
 		unsigned long cam2)
 {
-	settlbcam(0, PAGE_OFFSET, PPC_MEMSTART, cam0, _PAGE_KERNEL, 0);
+	settlbcam(0, PAGE_OFFSET, memstart_addr, cam0, _PAGE_KERNEL, 0);
 	tlbcam_index++;
 	if (cam1) {
 		tlbcam_index++;
-		settlbcam(1, PAGE_OFFSET+cam0, PPC_MEMSTART+cam0, cam1, _PAGE_KERNEL, 0);
+		settlbcam(1, PAGE_OFFSET+cam0, memstart_addr+cam0, cam1, _PAGE_KERNEL, 0);
 	}
 	if (cam2) {
 		tlbcam_index++;
-		settlbcam(2, PAGE_OFFSET+cam0+cam1, PPC_MEMSTART+cam0+cam1, cam2, _PAGE_KERNEL, 0);
+		settlbcam(2, PAGE_OFFSET+cam0+cam1, memstart_addr+cam0+cam1, cam2, _PAGE_KERNEL, 0);
 	}
 }
 
diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c
index 59a725b..01a81a0 100644
--- a/arch/powerpc/mm/init_32.c
+++ b/arch/powerpc/mm/init_32.c
@@ -59,8 +59,9 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 unsigned long total_memory;
 unsigned long total_lowmem;
 
-unsigned long ppc_memstart;
-unsigned long ppc_memoffset = PAGE_OFFSET;
+phys_addr_t memstart_addr;
+EXPORT_SYMBOL(memstart_addr);
+phys_addr_t lowmem_end_addr;
 
 int boot_mapsize;
 #ifdef CONFIG_PPC_PMAC
@@ -145,8 +146,7 @@ void __init MMU_init(void)
 		printk(KERN_WARNING "Only using first contiguous memory region");
 	}
 
-	total_memory = lmb_end_of_DRAM();
-	total_lowmem = total_memory;
+	total_lowmem = total_memory = lmb_end_of_DRAM() - memstart_addr;
 
 #ifdef CONFIG_FSL_BOOKE
 	/* Freescale Book-E parts expect lowmem to be mapped by fixed TLB
diff --git a/arch/powerpc/mm/mmu_decl.h b/arch/powerpc/mm/mmu_decl.h
index ebfd13d..5bc11f5 100644
--- a/arch/powerpc/mm/mmu_decl.h
+++ b/arch/powerpc/mm/mmu_decl.h
@@ -51,6 +51,7 @@ extern unsigned long __max_low_memory;
 extern unsigned long __initial_memory_limit;
 extern unsigned long total_memory;
 extern unsigned long total_lowmem;
+extern phys_addr_t memstart_addr;
 
 /* ...and now those things that may be slightly different between processor
  * architectures.  -- Dan
diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c
index ac3390f..64c44bc 100644
--- a/arch/powerpc/mm/pgtable_32.c
+++ b/arch/powerpc/mm/pgtable_32.c
@@ -281,12 +281,13 @@ int map_page(unsigned long va, phys_addr_t pa, int flags)
  */
 void __init mapin_ram(void)
 {
-	unsigned long v, p, s, f;
+	unsigned long v, s, f;
+	phys_addr_t p;
 	int ktext;
 
 	s = mmu_mapin_ram();
 	v = KERNELBASE + s;
-	p = PPC_MEMSTART + s;
+	p = memstart_addr + s;
 	for (; s < total_lowmem; s += PAGE_SIZE) {
 		ktext = ((char *) v >= _stext && (char *) v < etext);
 		f = ktext ?_PAGE_RAM_TEXT : _PAGE_RAM;
diff --git a/arch/powerpc/mm/ppc_mmu_32.c b/arch/powerpc/mm/ppc_mmu_32.c
index 72de3c7..65f915c 100644
--- a/arch/powerpc/mm/ppc_mmu_32.c
+++ b/arch/powerpc/mm/ppc_mmu_32.c
@@ -82,7 +82,6 @@ unsigned long __init mmu_mapin_ram(void)
 #else
 	unsigned long tot, bl, done;
 	unsigned long max_size = (256<<20);
-	unsigned long align;
 
 	if (__map_without_bats) {
 		printk(KERN_DEBUG "RAM mapped without BATs\n");
@@ -93,19 +92,13 @@ unsigned long __init mmu_mapin_ram(void)
 
 	/* Make sure we don't map a block larger than the
 	   smallest alignment of the physical address. */
-	/* alignment of PPC_MEMSTART */
-	align = ~(PPC_MEMSTART-1) & PPC_MEMSTART;
-	/* set BAT block size to MIN(max_size, align) */
-	if (align && align < max_size)
-		max_size = align;
-
 	tot = total_lowmem;
 	for (bl = 128<<10; bl < max_size; bl <<= 1) {
 		if (bl * 2 > tot)
 			break;
 	}
 
-	setbat(2, KERNELBASE, PPC_MEMSTART, bl, _PAGE_RAM);
+	setbat(2, KERNELBASE, 0, bl, _PAGE_RAM);
 	done = (unsigned long)bat_addrs[2].limit - KERNELBASE + 1;
 	if ((done < tot) && !bat_addrs[3].limit) {
 		/* use BAT3 to cover a bit more */
@@ -113,7 +106,7 @@ unsigned long __init mmu_mapin_ram(void)
 		for (bl = 128<<10; bl < max_size; bl <<= 1)
 			if (bl * 2 > tot)
 				break;
-		setbat(3, KERNELBASE+done, PPC_MEMSTART+done, bl, _PAGE_RAM);
+		setbat(3, KERNELBASE+done, done, bl, _PAGE_RAM);
 		done = (unsigned long)bat_addrs[3].limit - KERNELBASE + 1;
 	}
 
diff --git a/include/asm-powerpc/page_32.h b/include/asm-powerpc/page_32.h
index 65ea19e..51f8134 100644
--- a/include/asm-powerpc/page_32.h
+++ b/include/asm-powerpc/page_32.h
@@ -3,8 +3,6 @@
 
 #define VM_DATA_DEFAULT_FLAGS	VM_DATA_DEFAULT_FLAGS32
 
-#define PPC_MEMSTART	0
-
 #ifdef CONFIG_NOT_COHERENT_CACHE
 #define ARCH_KMALLOC_MINALIGN	L1_CACHE_BYTES
 #endif
-- 
1.5.4.1

[PATCH 05/11] [POWERPC] Introduce lowmem_end_addr to distiguish from total_lowmem

[Kumar Gala]

total_lowmem represents the amount of low memory not the physical address
that low memory ends at.  If the start of memory is at 0 it happends that
total_lowmem can be used as both the size and the address that lowmem
ends at. (technical its one byte beyond the end)

To make the code a bit more clear and deal with the case when the start of
memory isn't at physical 0, we introduce lowmem_end_addr that represents
one byte beyond the last physical address in the lowmem region.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/mm/44x_mmu.c  |    2 +-
 arch/powerpc/mm/init_32.c  |    4 +++-
 arch/powerpc/mm/init_64.c  |    2 ++
 arch/powerpc/mm/mem.c      |   16 +++++++++-------
 arch/powerpc/mm/mmu_decl.h |    1 +
 5 files changed, 16 insertions(+), 9 deletions(-)

diff --git a/arch/powerpc/mm/44x_mmu.c b/arch/powerpc/mm/44x_mmu.c
index 04dc087..953fb91 100644
--- a/arch/powerpc/mm/44x_mmu.c
+++ b/arch/powerpc/mm/44x_mmu.c
@@ -67,7 +67,7 @@ unsigned long __init mmu_mapin_ram(void)
 
 	/* Pin in enough TLBs to cover any lowmem not covered by the
 	 * initial 256M mapping established in head_44x.S */
-	for (addr = PPC_PIN_SIZE; addr < total_lowmem;
+	for (addr = PPC_PIN_SIZE; addr < lowmem_end_addr;
 	     addr += PPC_PIN_SIZE)
 		ppc44x_pin_tlb(addr + PAGE_OFFSET, addr);
 
diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c
index 01a81a0..345a275 100644
--- a/arch/powerpc/mm/init_32.c
+++ b/arch/powerpc/mm/init_32.c
@@ -147,6 +147,7 @@ void __init MMU_init(void)
 	}
 
 	total_lowmem = total_memory = lmb_end_of_DRAM() - memstart_addr;
+	lowmem_end_addr = memstart_addr + total_lowmem;
 
 #ifdef CONFIG_FSL_BOOKE
 	/* Freescale Book-E parts expect lowmem to be mapped by fixed TLB
@@ -157,9 +158,10 @@ void __init MMU_init(void)
 
 	if (total_lowmem > __max_low_memory) {
 		total_lowmem = __max_low_memory;
+		lowmem_end_addr = memstart_addr + total_lowmem;
 #ifndef CONFIG_HIGHMEM
 		total_memory = total_lowmem;
-		lmb_enforce_memory_limit(total_lowmem);
+		lmb_enforce_memory_limit(lowmem_end_addr);
 		lmb_analyze();
 #endif /* CONFIG_HIGHMEM */
 	}
diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c
index abeb0eb..f18b203 100644
--- a/arch/powerpc/mm/init_64.c
+++ b/arch/powerpc/mm/init_64.c
@@ -75,6 +75,8 @@
 /* max amount of RAM to use */
 unsigned long __max_memory;
 
+phys_addr_t memstart_addr;
+
 void free_initmem(void)
 {
 	unsigned long addr;
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 60c019c..9c10b14 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -217,9 +217,11 @@ void __init do_init_bootmem(void)
 	unsigned long total_pages;
 	int boot_mapsize;
 
-	max_pfn = total_pages = lmb_end_of_DRAM() >> PAGE_SHIFT;
+	max_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT;
+	total_pages = (lmb_end_of_DRAM() - memstart_addr) >> PAGE_SHIFT;
 #ifdef CONFIG_HIGHMEM
 	total_pages = total_lowmem >> PAGE_SHIFT;
+	max_low_pfn = lowmem_end_addr >> PAGE_SHIFT;
 #endif
 
 	/*
@@ -245,18 +247,18 @@ void __init do_init_bootmem(void)
 	 * present.
 	 */
 #ifdef CONFIG_HIGHMEM
-	free_bootmem_with_active_regions(0, total_lowmem >> PAGE_SHIFT);
+	free_bootmem_with_active_regions(0, lowmem_end_addr >> PAGE_SHIFT);
 
 	/* reserve the sections we're already using */
 	for (i = 0; i < lmb.reserved.cnt; i++) {
 		unsigned long addr = lmb.reserved.region[i].base +
 				     lmb_size_bytes(&lmb.reserved, i) - 1;
-		if (addr < total_lowmem)
+		if (addr < lowmem_end_addr)
 			reserve_bootmem(lmb.reserved.region[i].base,
 					lmb_size_bytes(&lmb.reserved, i),
 					BOOTMEM_DEFAULT);
-		else if (lmb.reserved.region[i].base < total_lowmem) {
-			unsigned long adjusted_size = total_lowmem -
+		else if (lmb.reserved.region[i].base < lowmem_end_addr) {
+			unsigned long adjusted_size = lowmem_end_addr -
 				      lmb.reserved.region[i].base;
 			reserve_bootmem(lmb.reserved.region[i].base,
 					adjusted_size, BOOTMEM_DEFAULT);
@@ -326,7 +328,7 @@ void __init paging_init(void)
 	       (top_of_ram - total_ram) >> 20);
 	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
 #ifdef CONFIG_HIGHMEM
-	max_zone_pfns[ZONE_DMA] = total_lowmem >> PAGE_SHIFT;
+	max_zone_pfns[ZONE_DMA] = lowmem_end_addr >> PAGE_SHIFT;
 	max_zone_pfns[ZONE_HIGHMEM] = top_of_ram >> PAGE_SHIFT;
 #else
 	max_zone_pfns[ZONE_DMA] = top_of_ram >> PAGE_SHIFT;
@@ -381,7 +383,7 @@ void __init mem_init(void)
 	{
 		unsigned long pfn, highmem_mapnr;
 
-		highmem_mapnr = total_lowmem >> PAGE_SHIFT;
+		highmem_mapnr = lowmem_end_addr >> PAGE_SHIFT;
 		for (pfn = highmem_mapnr; pfn < max_mapnr; ++pfn) {
 			struct page *page = pfn_to_page(pfn);
 			if (lmb_is_reserved(pfn << PAGE_SHIFT))
diff --git a/arch/powerpc/mm/mmu_decl.h b/arch/powerpc/mm/mmu_decl.h
index 5bc11f5..67477e7 100644
--- a/arch/powerpc/mm/mmu_decl.h
+++ b/arch/powerpc/mm/mmu_decl.h
@@ -52,6 +52,7 @@ extern unsigned long __initial_memory_limit;
 extern unsigned long total_memory;
 extern unsigned long total_lowmem;
 extern phys_addr_t memstart_addr;
+extern phys_addr_t lowmem_end_addr;
 
 /* ...and now those things that may be slightly different between processor
  * architectures.  -- Dan
-- 
1.5.4.1

[PATCH 06/11] [POWERPC] 85xx: Cleanup TLB initialization

[Kumar Gala]

* Determine the RPN we are running the kernel at runtime rather
  than using compile time constant for initial TLB

* Cleanup adjust_total_lowmem() to respect memstart_addr and
  be a bit more clear on variables that are sizes vs addresses.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/kernel/head_fsl_booke.S |   34 ++++++++++++++++++++++++------
 arch/powerpc/mm/fsl_booke_mmu.c      |   37 ++++++++++++++-------------------
 2 files changed, 43 insertions(+), 28 deletions(-)

diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
index d9cc2c2..9f40b3e 100644
--- a/arch/powerpc/kernel/head_fsl_booke.S
+++ b/arch/powerpc/kernel/head_fsl_booke.S
@@ -68,7 +68,9 @@ _ENTRY(_start);
 	mr	r29,r5
 	mr	r28,r6
 	mr	r27,r7
+	li	r25,0		/* phys kernel start (low) */
 	li	r24,0		/* CPU number */
+	li	r23,0		/* phys kernel start (high) */
 
 /* We try to not make any assumptions about how the boot loader
  * setup or used the TLBs.  We invalidate all mappings from the
@@ -167,7 +169,28 @@ skpinv:	addi	r6,r6,1				/* Increment */
 	mtspr	SPRN_MAS0,r7
 	tlbre
 
-	/* Just modify the entry ID, EPN and RPN for the temp mapping */
+	/* grab and fixup the RPN */
+	mfspr	r6,SPRN_MAS1	/* extract MAS1[SIZE] */
+	rlwinm	r6,r6,25,27,30
+	li	r8,-1
+	addi	r6,r6,10
+	slw	r6,r8,r6	/* convert to mask */
+
+	bl	1f		/* Find our address */
+1:	mflr	r7
+
+	mfspr	r8,SPRN_MAS3
+#ifdef CONFIG_PHYS_64BIT
+	mfspr	r23,SPRN_MAS7
+#endif
+	and	r8,r6,r8
+	subfic	r9,r6,-4096
+	and	r9,r9,r7
+
+	or	r25,r8,r9
+	ori	r8,r25,(MAS3_SX|MAS3_SW|MAS3_SR)
+
+	/* Just modify the entry ID and EPN for the temp mapping */
 	lis	r7,0x1000	/* Set MAS0(TLBSEL) = 1 */
 	rlwimi	r7,r5,16,4,15	/* Setup MAS0 = TLBSEL | ESEL(r5) */
 	mtspr	SPRN_MAS0,r7
@@ -177,12 +200,10 @@ skpinv:	addi	r6,r6,1				/* Increment */
 	ori	r6,r6,(MAS1_TSIZE(BOOKE_PAGESZ_4K))@l
 	mtspr	SPRN_MAS1,r6
 	mfspr	r6,SPRN_MAS2
-	lis	r7,PHYSICAL_START@h
+	li	r7,0		/* temp EPN = 0 */
 	rlwimi	r7,r6,0,20,31
 	mtspr	SPRN_MAS2,r7
-	mfspr	r6,SPRN_MAS3
-	rlwimi	r7,r6,0,20,31
-	mtspr	SPRN_MAS3,r7
+	mtspr	SPRN_MAS3,r8
 	tlbwe
 
 	xori	r6,r4,1
@@ -232,8 +253,7 @@ skpinv:	addi	r6,r6,1				/* Increment */
 	ori	r6,r6,PAGE_OFFSET@l
 	rlwimi	r6,r7,0,20,31
 	mtspr	SPRN_MAS2,r6
-	li	r7,(MAS3_SX|MAS3_SW|MAS3_SR)
-	mtspr	SPRN_MAS3,r7
+	mtspr	SPRN_MAS3,r8
 	tlbwe
 
 /* 7. Jump to KERNELBASE mapping */
diff --git a/arch/powerpc/mm/fsl_booke_mmu.c b/arch/powerpc/mm/fsl_booke_mmu.c
index 3dd0c81..59f6649 100644
--- a/arch/powerpc/mm/fsl_booke_mmu.c
+++ b/arch/powerpc/mm/fsl_booke_mmu.c
@@ -49,7 +49,6 @@
 #include <asm/mmu.h>
 #include <asm/uaccess.h>
 #include <asm/smp.h>
-#include <asm/bootx.h>
 #include <asm/machdep.h>
 #include <asm/setup.h>
 
@@ -59,7 +58,6 @@ extern void loadcam_entry(unsigned int index);
 unsigned int tlbcam_index;
 unsigned int num_tlbcam_entries;
 static unsigned long __cam0, __cam1, __cam2;
-#define MAX_LOW_MEM	CONFIG_LOWMEM_SIZE
 
 #define NUM_TLBCAMS	(16)
 
@@ -195,35 +193,32 @@ unsigned long __init mmu_mapin_ram(void)
 void __init
 adjust_total_lowmem(void)
 {
-	unsigned long max_low_mem = MAX_LOW_MEM;
-	unsigned long cam_max = 0x10000000;
-	unsigned long ram;
+	phys_addr_t max_lowmem_size = __max_low_memory;
+	phys_addr_t cam_max_size = 0x10000000;
+	phys_addr_t ram;
 
-	/* adjust CAM size to max_low_mem */
-	if (max_low_mem < cam_max)
-		cam_max = max_low_mem;
+	/* adjust CAM size to max_lowmem_size */
+	if (max_lowmem_size < cam_max_size)
+		cam_max_size = max_lowmem_size;
 
-	/* adjust lowmem size to max_low_mem */
-	if (max_low_mem < total_lowmem)
-		ram = max_low_mem;
-	else
-		ram = total_lowmem;
+	/* adjust lowmem size to max_lowmem_size */
+	ram = min(max_lowmem_size, total_lowmem);
 
 	/* Calculate CAM values */
 	__cam0 = 1UL << 2 * (__ilog2(ram) / 2);
-	if (__cam0 > cam_max)
-		__cam0 = cam_max;
+	if (__cam0 > cam_max_size)
+		__cam0 = cam_max_size;
 	ram -= __cam0;
 	if (ram) {
 		__cam1 = 1UL << 2 * (__ilog2(ram) / 2);
-		if (__cam1 > cam_max)
-			__cam1 = cam_max;
+		if (__cam1 > cam_max_size)
+			__cam1 = cam_max_size;
 		ram -= __cam1;
 	}
 	if (ram) {
 		__cam2 = 1UL << 2 * (__ilog2(ram) / 2);
-		if (__cam2 > cam_max)
-			__cam2 = cam_max;
+		if (__cam2 > cam_max_size)
+			__cam2 = cam_max_size;
 		ram -= __cam2;
 	}
 
@@ -231,6 +226,6 @@ adjust_total_lowmem(void)
 			" CAM2=%ldMb residual: %ldMb\n",
 			__cam0 >> 20, __cam1 >> 20, __cam2 >> 20,
 			(total_lowmem - __cam0 - __cam1 - __cam2) >> 20);
-	__max_low_memory = max_low_mem = __cam0 + __cam1 + __cam2;
-	__initial_memory_limit = __max_low_memory;
+	__max_low_memory = __cam0 + __cam1 + __cam2;
+	__initial_memory_limit = memstart_addr + __max_low_memory;
 }
-- 
1.5.4.1

[PATCH 07/11] [POWERPC] Use lowmem_end_addr to limit lmb allocations on ppc32

[Kumar Gala]

Now that we have a proper variable that is the address of the top
of low memory we can use it to limit the lmb allocations.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 include/asm-powerpc/lmb.h |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/asm-powerpc/lmb.h b/include/asm-powerpc/lmb.h
index 028184b..6f5fdf0 100644
--- a/include/asm-powerpc/lmb.h
+++ b/include/asm-powerpc/lmb.h
@@ -6,8 +6,8 @@
 #define LMB_DBG(fmt...) udbg_printf(fmt)
 
 #ifdef CONFIG_PPC32
-extern unsigned long __max_low_memory;
-#define LMB_REAL_LIMIT	__max_low_memory
+extern phys_addr_t lowmem_end_addr;
+#define LMB_REAL_LIMIT	lowmem_end_addr
 #else
 #define LMB_REAL_LIMIT	0
 #endif
-- 
1.5.4.1

[PATCH 08/11 v2] [POWERPC] Rename __initial_memory_limit to __initial_memory_limit_addr

[Kumar Gala]

We always use __initial_memory_limit as an address so rename it
to be clear.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---

Fixed a build breakage issue on ebony due to phys_addr_t not being the type
of __initial_memory_limit_addr.

 arch/powerpc/mm/fsl_booke_mmu.c |    2 +-
 arch/powerpc/mm/init_32.c       |   10 +++++-----
 arch/powerpc/mm/mmu_decl.h      |    2 +-
 arch/powerpc/mm/ppc_mmu_32.c    |    2 +-
 4 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/arch/powerpc/mm/fsl_booke_mmu.c b/arch/powerpc/mm/fsl_booke_mmu.c
index 59f6649..ada249b 100644
--- a/arch/powerpc/mm/fsl_booke_mmu.c
+++ b/arch/powerpc/mm/fsl_booke_mmu.c
@@ -227,5 +227,5 @@ adjust_total_lowmem(void)
 			__cam0 >> 20, __cam1 >> 20, __cam2 >> 20,
 			(total_lowmem - __cam0 - __cam1 - __cam2) >> 20);
 	__max_low_memory = __cam0 + __cam1 + __cam2;
-	__initial_memory_limit = memstart_addr + __max_low_memory;
+	__initial_memory_limit_addr = memstart_addr + __max_low_memory;
 }
diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c
index 345a275..2c72c39 100644
--- a/arch/powerpc/mm/init_32.c
+++ b/arch/powerpc/mm/init_32.c
@@ -96,10 +96,10 @@ int __map_without_ltlbs;
 unsigned long __max_low_memory = MAX_LOW_MEM;
 
 /*
- * limit of what is accessible with initial MMU setup -
+ * address of the limit of what is accessible with initial MMU setup -
  * 256MB usually, but only 16MB on 601.
  */
-unsigned long __initial_memory_limit = 0x10000000;
+phys_addr_t __initial_memory_limit_addr = (phys_addr_t)0x10000000;
 
 /*
  * Check for command-line options that affect what MMU_init will do.
@@ -132,10 +132,10 @@ void __init MMU_init(void)
 
 	/* 601 can only access 16MB at the moment */
 	if (PVR_VER(mfspr(SPRN_PVR)) == 1)
-		__initial_memory_limit = 0x01000000;
+		__initial_memory_limit_addr = 0x01000000;
 	/* 8xx can only access 8MB at the moment */
 	if (PVR_VER(mfspr(SPRN_PVR)) == 0x50)
-		__initial_memory_limit = 0x00800000;
+		__initial_memory_limit_addr = 0x00800000;
 
 	/* parse args from command line */
 	MMU_setup();
@@ -210,7 +210,7 @@ void __init *early_get_page(void)
 		p = alloc_bootmem_pages(PAGE_SIZE);
 	} else {
 		p = __va(lmb_alloc_base(PAGE_SIZE, PAGE_SIZE,
-					__initial_memory_limit));
+					__initial_memory_limit_addr));
 	}
 	return p;
 }
diff --git a/arch/powerpc/mm/mmu_decl.h b/arch/powerpc/mm/mmu_decl.h
index 67477e7..0480225 100644
--- a/arch/powerpc/mm/mmu_decl.h
+++ b/arch/powerpc/mm/mmu_decl.h
@@ -48,7 +48,7 @@ extern unsigned int num_tlbcam_entries;
 
 extern unsigned long ioremap_bot;
 extern unsigned long __max_low_memory;
-extern unsigned long __initial_memory_limit;
+extern phys_addr_t __initial_memory_limit_addr;
 extern unsigned long total_memory;
 extern unsigned long total_lowmem;
 extern phys_addr_t memstart_addr;
diff --git a/arch/powerpc/mm/ppc_mmu_32.c b/arch/powerpc/mm/ppc_mmu_32.c
index 65f915c..cef9f15 100644
--- a/arch/powerpc/mm/ppc_mmu_32.c
+++ b/arch/powerpc/mm/ppc_mmu_32.c
@@ -233,7 +233,7 @@ void __init MMU_init_hw(void)
 	 */
 	if ( ppc_md.progress ) ppc_md.progress("hash:find piece", 0x322);
 	Hash = __va(lmb_alloc_base(Hash_size, Hash_size,
-				   __initial_memory_limit));
+				   __initial_memory_limit_addr));
 	cacheable_memzero(Hash, Hash_size);
 	_SDR1 = __pa(Hash) | SDR1_LOW_BITS;
 
-- 
1.5.4.1

[PATCH 09/11] [POWERPC] Clean up some linker and symbol usage

[Kumar Gala]

* use _stext and _end sybmols when reserving kernel text in the lmb.  Use
  of these symbols is a bit more robust way to determine the physical start
  and size of the kernel text.
* PAGE_OFFSET is not always the start of code, use _stext instead.
* grab PAGE_SIZE and KERNELBASE from asm/page.h like ppc64 does.  Makes the
  code a bit more common and provide a single place to manipulate the
  defines for things like kdump.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/kernel/prom.c        |    2 +-
 arch/powerpc/kernel/setup_32.c    |    2 +-
 arch/powerpc/kernel/setup_64.c    |    2 +-
 arch/powerpc/kernel/vmlinux.lds.S |    4 +---
 4 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 9330920..60ef7d1 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -1124,7 +1124,7 @@ void __init early_init_devtree(void *params)
 	parse_early_param();
 
 	/* Reserve LMB regions used by kernel, initrd, dt, etc... */
-	lmb_reserve(PHYSICAL_START, __pa(klimit) - PHYSICAL_START);
+	lmb_reserve(__pa(_stext), _end - _stext);
 	reserve_kdump_trampoline();
 	reserve_crashkernel();
 	early_reserve_mem();
diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index cd870a8..b0989ca 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -277,7 +277,7 @@ void __init setup_arch(char **cmdline_p)
 	if (ppc_md.panic)
 		setup_panic();
 
-	init_mm.start_code = PAGE_OFFSET;
+	init_mm.start_code = (unsigned long)_stext;
 	init_mm.end_code = (unsigned long) _etext;
 	init_mm.end_data = (unsigned long) _edata;
 	init_mm.brk = klimit;
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 2c2d831..0205d40 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -510,7 +510,7 @@ void __init setup_arch(char **cmdline_p)
 	if (ppc_md.panic)
 		setup_panic();
 
-	init_mm.start_code = PAGE_OFFSET;
+	init_mm.start_code = (unsigned long)_stext;
 	init_mm.end_code = (unsigned long) _etext;
 	init_mm.end_data = (unsigned long) _edata;
 	init_mm.brk = klimit;
diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S
index 0afb9e3..b5a76bc 100644
--- a/arch/powerpc/kernel/vmlinux.lds.S
+++ b/arch/powerpc/kernel/vmlinux.lds.S
@@ -1,11 +1,9 @@
 #ifdef CONFIG_PPC64
-#include <asm/page.h>
 #define PROVIDE32(x)	PROVIDE(__unused__##x)
 #else
-#define PAGE_SIZE	4096
-#define KERNELBASE	CONFIG_KERNEL_START
 #define PROVIDE32(x)	PROVIDE(x)
 #endif
+#include <asm/page.h>
 #include <asm-generic/vmlinux.lds.h>
 #include <asm/cache.h>
 
-- 
1.5.4.1

[PATCH 10/11] [POWERPC] Move phys_addr_t definition into asm/types.h

[Kumar Gala]

Moved phys_addr_t out of mmu-*.h and into asm/types.h so we can use it in
places that before would have caused recursive includes.

For example to use phys_addr_t in <asm/page.h> we would have included
<asm/mmu.h> which would have possibly included <asm/mmu-hash64.h> which
includes <asm/page.h>.  Wheeee recursive include.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 include/asm-powerpc/mmu-40x.h       |    2 --
 include/asm-powerpc/mmu-44x.h       |    2 --
 include/asm-powerpc/mmu-8xx.h       |    2 --
 include/asm-powerpc/mmu-fsl-booke.h |    6 ------
 include/asm-powerpc/mmu-hash32.h    |    2 --
 include/asm-powerpc/mmu-hash64.h    |    3 ---
 include/asm-powerpc/types.h         |    7 +++++++
 7 files changed, 7 insertions(+), 17 deletions(-)

diff --git a/include/asm-powerpc/mmu-40x.h b/include/asm-powerpc/mmu-40x.h
index 7d37f77..3d10867 100644
--- a/include/asm-powerpc/mmu-40x.h
+++ b/include/asm-powerpc/mmu-40x.h
@@ -53,8 +53,6 @@
 
 #ifndef __ASSEMBLY__
 
-typedef unsigned long phys_addr_t;
-
 typedef struct {
 	unsigned long id;
 	unsigned long vdso_base;
diff --git a/include/asm-powerpc/mmu-44x.h b/include/asm-powerpc/mmu-44x.h
index 62772ae..c8b02d9 100644
--- a/include/asm-powerpc/mmu-44x.h
+++ b/include/asm-powerpc/mmu-44x.h
@@ -53,8 +53,6 @@
 
 #ifndef __ASSEMBLY__
 
-typedef unsigned long long phys_addr_t;
-
 typedef struct {
 	unsigned long id;
 	unsigned long vdso_base;
diff --git a/include/asm-powerpc/mmu-8xx.h b/include/asm-powerpc/mmu-8xx.h
index 952bd88..9db877e 100644
--- a/include/asm-powerpc/mmu-8xx.h
+++ b/include/asm-powerpc/mmu-8xx.h
@@ -136,8 +136,6 @@
 #define SPRN_M_TW	799
 
 #ifndef __ASSEMBLY__
-typedef unsigned long phys_addr_t;
-
 typedef struct {
 	unsigned long id;
 	unsigned long vdso_base;
diff --git a/include/asm-powerpc/mmu-fsl-booke.h b/include/asm-powerpc/mmu-fsl-booke.h
index 3758000..925d93c 100644
--- a/include/asm-powerpc/mmu-fsl-booke.h
+++ b/include/asm-powerpc/mmu-fsl-booke.h
@@ -73,12 +73,6 @@
 
 #ifndef __ASSEMBLY__
 
-#ifndef CONFIG_PHYS_64BIT
-typedef unsigned long phys_addr_t;
-#else
-typedef unsigned long long phys_addr_t;
-#endif
-
 typedef struct {
 	unsigned long id;
 	unsigned long vdso_base;
diff --git a/include/asm-powerpc/mmu-hash32.h b/include/asm-powerpc/mmu-hash32.h
index 4bd735b..6e21ca6 100644
--- a/include/asm-powerpc/mmu-hash32.h
+++ b/include/asm-powerpc/mmu-hash32.h
@@ -84,8 +84,6 @@ typedef struct {
 	unsigned long vdso_base;
 } mm_context_t;
 
-typedef unsigned long phys_addr_t;
-
 #endif /* !__ASSEMBLY__ */
 
 #endif /* _ASM_POWERPC_MMU_HASH32_H_ */
diff --git a/include/asm-powerpc/mmu-hash64.h b/include/asm-powerpc/mmu-hash64.h
index 2864fa3..0dff767 100644
--- a/include/asm-powerpc/mmu-hash64.h
+++ b/include/asm-powerpc/mmu-hash64.h
@@ -469,9 +469,6 @@ static inline unsigned long get_vsid(unsigned long context, unsigned long ea,
 				 VSID_MODULUS_256M)
 #define KERNEL_VSID(ea)		VSID_SCRAMBLE(GET_ESID(ea))
 
-/* Physical address used by some IO functions */
-typedef unsigned long phys_addr_t;
-
 #endif /* __ASSEMBLY__ */
 
 #endif /* _ASM_POWERPC_MMU_HASH64_H_ */
diff --git a/include/asm-powerpc/types.h b/include/asm-powerpc/types.h
index 903fd19..020db52 100644
--- a/include/asm-powerpc/types.h
+++ b/include/asm-powerpc/types.h
@@ -50,6 +50,13 @@ typedef struct {
 	__u32 u[4];
 } __attribute__((aligned(16))) __vector128;
 
+/* Physical address used by some IO functions */
+#ifndef CONFIG_PHYS_64BIT
+typedef unsigned long phys_addr_t;
+#else
+typedef unsigned long long phys_addr_t;
+#endif
+
 #endif /* __ASSEMBLY__ */
 
 #ifdef __KERNEL__
-- 
1.5.4.1

[PATCH 11/11] [POWERPC] 85xx: Add support for relocatble kernel (and booting at non-zero)

[Kumar Gala]

Added support to allow an 85xx kernel to be run from a non-zero physical
address (useful for cooperative asymmetric multiprocessing situations) and
kdump.  The support can either be at compile time or runtime
(CONFIG_RELOCATABLE).

Currently we are limited to running at a physical address that is module
256M.  This is due to how we map TLBs to cover lowmem and should be fixed
up to allow 64M or maybe even 16M alignment in the future.

All the magic for this support is accomplished by proper initializating
of the kernel memory subsystem properly and ARCH_PFN_OFFSET.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/Kconfig                 |   69 ++++++++++++++++++++++++++++++++-
 arch/powerpc/boot/Makefile           |    4 +-
 arch/powerpc/kernel/head_fsl_booke.S |   11 +++++
 arch/powerpc/kernel/prom.c           |    4 ++
 arch/powerpc/kernel/setup_64.c       |    2 +-
 arch/powerpc/mm/init_32.c            |    4 +-
 arch/powerpc/mm/init_64.c            |    3 +-
 arch/powerpc/mm/mem.c                |    5 +-
 include/asm-powerpc/kdump.h          |    5 --
 include/asm-powerpc/page.h           |   43 +++++++++++++++++---
 include/asm-powerpc/page_32.h        |    6 +++
 include/asm-powerpc/pgtable-ppc32.h  |    5 +--
 12 files changed, 135 insertions(+), 26 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 4d9ced2..42c22f7 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -619,21 +619,76 @@ config LOWMEM_SIZE
 	hex "Maximum low memory size (in bytes)" if LOWMEM_SIZE_BOOL
 	default "0x30000000"
 
+config RELOCATABLE
+	bool "Build a relocatable kernel (EXPERIMENTAL)"
+	depends on EXPERIMENTAL && ADVANCED_OPTIONS && FLATMEM && FSL_BOOKE
+	help
+	  This builds a kernel image that is capable of running at the
+	  location the kernel is loaded at (some alignment restrictions may
+	  exist).
+
+	  One use is for the kexec on panic case where the recovery kernel
+	  must live at a different physical address than the primary
+	  kernel.
+
+	  Note: If CONFIG_RELOCATABLE=y, then the kernel runs from the address
+	  it has been loaded at and the compile time physical addresses
+	  CONFIG_PHYSICAL_START is ignored.  However CONFIG_PHYSICAL_START
+	  setting can still be useful to bootwrappers that need to know the
+	  load location of the kernel (eg. u-boot/mkimage).
+
+config PAGE_OFFSET_BOOL
+	bool "Set custom page offset address"
+	depends on ADVANCED_OPTIONS
+	help
+	  This option allows you to set the kernel virtual address at which
+	  the kernel will map low memory.  This can be useful in optimizing
+	  the virtual memory layout of the system.
+
+	  Say N here unless you know what you are doing.
+
+config PAGE_OFFSET
+	hex "Virtual address of memory base" if PAGE_OFFSET_BOOL
+	default "0xc0000000"
+
 config KERNEL_START_BOOL
 	bool "Set custom kernel base address"
 	depends on ADVANCED_OPTIONS
 	help
 	  This option allows you to set the kernel virtual address at which
-	  the kernel will map low memory (the kernel image will be linked at
-	  this address).  This can be useful in optimizing the virtual memory
-	  layout of the system.
+	  the kernel will be loaded.  Normally this should match PAGE_OFFSET
+	  however there are times (like kdump) that one might not want them
+	  to be the same.
 
 	  Say N here unless you know what you are doing.
 
 config KERNEL_START
 	hex "Virtual address of kernel base" if KERNEL_START_BOOL
+	default PAGE_OFFSET if PAGE_OFFSET_BOOL
+	default "0xc2000000" if CRASH_DUMP
 	default "0xc0000000"
 
+config PHYSICAL_START_BOOL
+	bool "Set physical address where the kernel is loaded"
+	depends on ADVANCED_OPTIONS && FLATMEM && FSL_BOOKE
+	help
+	  This gives the physical address where the kernel is loaded.
+
+	  Say N here unless you know what you are doing.
+
+config PHYSICAL_START
+	hex "Physical address where the kernel is loaded" if PHYSICAL_START_BOOL
+	default "0x02000000" if PPC_STD_MMU && CRASH_DUMP
+	default "0x00000000"
+
+config PHYSICAL_ALIGN
+	hex
+	default "0x10000000" if FSL_BOOKE
+	help
+	  This value puts the alignment restrictions on physical address
+	  where kernel is loaded and run from. Kernel is compiled for an
+	  address which meets above alignment restriction.
+
 config TASK_SIZE_BOOL
 	bool "Set custom user task size"
 	depends on ADVANCED_OPTIONS
@@ -680,9 +735,17 @@ config PIN_TLB
 endmenu
 
 if PPC64
+config PAGE_OFFSET
+	hex
+	default "0xc000000000000000"
 config KERNEL_START
 	hex
+	default "0xc000000002000000" if CRASH_DUMP
 	default "0xc000000000000000"
+config PHYSICAL_START
+	hex
+	default "0x02000000" if CRASH_DUMP
+	default "0x00000000"
 endif
 
 source "net/Kconfig"
diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile
index 3c80858..c7c2a9d 100644
--- a/arch/powerpc/boot/Makefile
+++ b/arch/powerpc/boot/Makefile
@@ -35,8 +35,8 @@ endif
 
 BOOTCFLAGS	+= -I$(obj) -I$(srctree)/$(obj) -I$(srctree)/$(src)/libfdt
 
-ifdef CONFIG_MEMORY_START
-MEMBASE=$(CONFIG_MEMORY_START)
+ifdef CONFIG_PHYSICAL_START
+MEMBASE=$(CONFIG_PHYSICAL_START)
 else
 MEMBASE=0x00000000
 endif
diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
index 9f40b3e..4d0336b 100644
--- a/arch/powerpc/kernel/head_fsl_booke.S
+++ b/arch/powerpc/kernel/head_fsl_booke.S
@@ -368,6 +368,17 @@ skpinv:	addi	r6,r6,1				/* Increment */
 
 	bl	early_init
 
+#ifdef CONFIG_RELOCATABLE
+	lis	r3,kernstart_addr@ha
+	la	r3,kernstart_addr@l(r3)
+#ifdef CONFIG_PHYS_64BIT
+	stw	r23,0(r3)
+	stw	r25,4(r3)
+#else
+	stw	r25,0(r3)
+#endif
+#endif
+
 	mfspr	r3,SPRN_TLB1CFG
 	andi.	r3,r3,0xfff
 	lis	r4,num_tlbcam_entries@ha
diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 60ef7d1..988cbde 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -53,6 +53,7 @@
 #include <asm/pci-bridge.h>
 #include <asm/phyp_dump.h>
 #include <asm/kexec.h>
+#include <mm/mmu_decl.h>
 
 #ifdef DEBUG
 #define DBG(fmt...) printk(KERN_ERR fmt)
@@ -978,7 +979,10 @@ static int __init early_init_dt_scan_memory(unsigned long node,
 		}
 #endif
 		lmb_add(base, size);
+
+		memstart_addr = min((u64)memstart_addr, base);
 	}
+
 	return 0;
 }
 
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 0205d40..9087e7a 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -431,7 +431,7 @@ void __init setup_system(void)
 		printk("htab_address                  = 0x%p\n", htab_address);
 	printk("htab_hash_mask                = 0x%lx\n", htab_hash_mask);
 #if PHYSICAL_START > 0
-	printk("physical_start                = 0x%x\n", PHYSICAL_START);
+	printk("physical_start                = 0x%lx\n", PHYSICAL_START);
 #endif
 	printk("-----------------------------------------------------\n");
 
diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c
index 2c72c39..4b97daa 100644
--- a/arch/powerpc/mm/init_32.c
+++ b/arch/powerpc/mm/init_32.c
@@ -59,8 +59,10 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 unsigned long total_memory;
 unsigned long total_lowmem;
 
-phys_addr_t memstart_addr;
+phys_addr_t memstart_addr = (phys_addr_t)~0ull;
 EXPORT_SYMBOL(memstart_addr);
+phys_addr_t kernstart_addr;
+EXPORT_SYMBOL(kernstart_addr);
 phys_addr_t lowmem_end_addr;
 
 int boot_mapsize;
diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c
index f18b203..7fbbafd 100644
--- a/arch/powerpc/mm/init_64.c
+++ b/arch/powerpc/mm/init_64.c
@@ -75,7 +75,8 @@
 /* max amount of RAM to use */
 unsigned long __max_memory;
 
-phys_addr_t memstart_addr;
+phys_addr_t memstart_addr = ~0;
+phys_addr_t kernstart_addr;
 
 void free_initmem(void)
 {
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 9c10b14..3ec7814 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -217,7 +217,7 @@ void __init do_init_bootmem(void)
 	unsigned long total_pages;
 	int boot_mapsize;
 
-	max_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT;
+	max_low_pfn = max_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT;
 	total_pages = (lmb_end_of_DRAM() - memstart_addr) >> PAGE_SHIFT;
 #ifdef CONFIG_HIGHMEM
 	total_pages = total_lowmem >> PAGE_SHIFT;
@@ -233,7 +233,8 @@ void __init do_init_bootmem(void)
 
 	start = lmb_alloc(bootmap_pages << PAGE_SHIFT, PAGE_SIZE);
 
-	boot_mapsize = init_bootmem(start >> PAGE_SHIFT, total_pages);
+	min_low_pfn = MEMORY_START >> PAGE_SHIFT;
+	boot_mapsize = init_bootmem_node(NODE_DATA(0), start >> PAGE_SHIFT, min_low_pfn, max_low_pfn);
 
 	/* Add active regions with valid PFNs */
 	for (i = 0; i < lmb.memory.cnt; i++) {
diff --git a/include/asm-powerpc/kdump.h b/include/asm-powerpc/kdump.h
index 10e8eb1..f6c93c7 100644
--- a/include/asm-powerpc/kdump.h
+++ b/include/asm-powerpc/kdump.h
@@ -11,16 +11,11 @@
 
 #ifdef CONFIG_CRASH_DUMP
 
-#define PHYSICAL_START	KDUMP_KERNELBASE
 #define KDUMP_TRAMPOLINE_START	0x0100
 #define KDUMP_TRAMPOLINE_END	0x3000
 
 #define KDUMP_MIN_TCE_ENTRIES	2048
 
-#else /* !CONFIG_CRASH_DUMP */
-
-#define PHYSICAL_START	0x0
-
 #endif /* CONFIG_CRASH_DUMP */
 
 #ifndef __ASSEMBLY__
diff --git a/include/asm-powerpc/page.h b/include/asm-powerpc/page.h
index df47bbb..adf0591 100644
--- a/include/asm-powerpc/page.h
+++ b/include/asm-powerpc/page.h
@@ -12,6 +12,7 @@
 
 #include <asm/asm-compat.h>
 #include <asm/kdump.h>
+#include <asm/types.h>
 
 /*
  * On PPC32 page size is 4K. For PPC64 we support either 4K or 64K software
@@ -42,8 +43,23 @@
  *
  * The kdump dump kernel is one example where KERNELBASE != PAGE_OFFSET.
  *
- * To get a physical address from a virtual one you subtract PAGE_OFFSET,
- * _not_ KERNELBASE.
+ * PAGE_OFFSET is the virtual address of the start of lowmem.
+ *
+ * PHYSICAL_START is the physical address of the start of the kernel.
+ *
+ * MEMORY_START is the physical address of the start of lowmem.
+ *
+ * KERNELBASE, PAGE_OFFSET, and PHYSICAL_START are all configurable on
+ * ppc32 and based on how they are set we determine MEMORY_START.
+ *
+ * For the linear mapping the following equation should be true:
+ * KERNELBASE - PAGE_OFFSET = PHYSICAL_START - MEMORY_START
+ *
+ * Also, KERNELBASE >= PAGE_OFFSET and PHYSICAL_START >= MEMORY_START
+ *
+ * There are two was to determine a physical address from a virtual one:
+ * va = pa + PAGE_OFFSET - MEMORY_START
+ * va = pa + KERNELBASE - PHYSICAL_START
  *
  * If you want to know something's offset from the start of the kernel you
  * should subtract KERNELBASE.
@@ -51,19 +67,32 @@
  * If you want to test if something's a kernel address, use is_kernel_addr().
  */
 
-#define PAGE_OFFSET     ASM_CONST(CONFIG_KERNEL_START)
-#define KERNELBASE      (PAGE_OFFSET + PHYSICAL_START)
+#define KERNELBASE      ASM_CONST(CONFIG_KERNEL_START)
+#define PAGE_OFFSET	ASM_CONST(CONFIG_PAGE_OFFSET)
+
+#if defined(CONFIG_RELOCATABLE) && defined(CONFIG_FLATMEM)
+#ifndef __ASSEMBLY__
+extern phys_addr_t memstart_addr;
+extern phys_addr_t kernstart_addr;
+#endif
+#define PHYSICAL_START	kernstart_addr
+#define MEMORY_START	memstart_addr
+#else
+#define PHYSICAL_START	ASM_CONST(CONFIG_PHYSICAL_START)
+#define MEMORY_START	(PHYSICAL_START + PAGE_OFFSET - KERNELBASE)
+#endif
 
 #ifdef CONFIG_FLATMEM
-#define pfn_valid(pfn)		((pfn) < max_mapnr)
+#define ARCH_PFN_OFFSET		(MEMORY_START >> PAGE_SHIFT)
+#define pfn_valid(pfn)		((pfn) >= ARCH_PFN_OFFSET && (pfn) < (ARCH_PFN_OFFSET + max_mapnr))
 #endif
 
 #define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
 #define pfn_to_kaddr(pfn)	__va((pfn) << PAGE_SHIFT)
 #define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
 
-#define __va(x) ((void *)((unsigned long)(x) + PAGE_OFFSET))
-#define __pa(x) ((unsigned long)(x) - PAGE_OFFSET)
+#define __va(x) ((void *)((unsigned long)(x) - PHYSICAL_START + KERNELBASE))
+#define __pa(x) ((unsigned long)(x) + PHYSICAL_START - KERNELBASE)
 
 /*
  * Unfortunately the PLT is in the BSS in the PPC32 ELF ABI,
diff --git a/include/asm-powerpc/page_32.h b/include/asm-powerpc/page_32.h
index 51f8134..ebfae53 100644
--- a/include/asm-powerpc/page_32.h
+++ b/include/asm-powerpc/page_32.h
@@ -1,6 +1,12 @@
 #ifndef _ASM_POWERPC_PAGE_32_H
 #define _ASM_POWERPC_PAGE_32_H
 
+#if defined(CONFIG_PHYSICAL_ALIGN) && (CONFIG_PHYSICAL_START != 0)
+#if (CONFIG_PHYSICAL_START % CONFIG_PHYSICAL_ALIGN) != 0
+#error "CONFIG_PHYSICAL_START must be a multiple of CONFIG_PHYSICAL_ALIGN"
+#endif
+#endif
+
 #define VM_DATA_DEFAULT_FLAGS	VM_DATA_DEFAULT_FLAGS32
 
 #ifdef CONFIG_NOT_COHERENT_CACHE
diff --git a/include/asm-powerpc/pgtable-ppc32.h b/include/asm-powerpc/pgtable-ppc32.h
index 2c79f55..dbd1875 100644
--- a/include/asm-powerpc/pgtable-ppc32.h
+++ b/include/asm-powerpc/pgtable-ppc32.h
@@ -98,9 +98,6 @@ extern int icache_44x_need_flush;
 #define USER_PTRS_PER_PGD	(TASK_SIZE / PGDIR_SIZE)
 #define FIRST_USER_ADDRESS	0
 
-#define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT)
-#define KERNEL_PGD_PTRS (PTRS_PER_PGD-USER_PGD_PTRS)
-
 #define pte_ERROR(e) \
 	printk("%s:%d: bad pte %llx.\n", __FILE__, __LINE__, \
 		(unsigned long long)pte_val(e))
@@ -692,7 +689,7 @@ extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
 #define pmd_page_vaddr(pmd)	\
 	((unsigned long) (pmd_val(pmd) & PAGE_MASK))
 #define pmd_page(pmd)		\
-	(mem_map + (__pa(pmd_val(pmd)) >> PAGE_SHIFT))
+	(mem_map + (__pa(pmd_val(pmd)) >> PAGE_SHIFT) - ARCH_PFN_OFFSET)
 #endif
 
 /* to find an entry in a kernel page-table-directory */
-- 
1.5.4.1

Re: [PATCH 04/11 v2] [POWERPC] Remove and replace uses of PPC_MEMSTART with memstart_addr

[Paul Mackerras]

Kumar Gala writes:

> For the sub-arches that support relocatable interrupt vectors (book-e) its
> reasonable to have memory start at a non-zero physical address.  For those
> cases use the variable memstart_addr instead of the #define PPC_MEMSTART
> since the only uses of PPC_MEMSTART are for initialization and in the
> future we can set memstart_addr at runtime to have a relocatable kernel.

In those cases, is it still true that the kernel sits at the start of
the usable memory, or might there be usable memory before _stext?
In other words, is PAGE_OFFSET == KERNELBASE still true or not?

Paul.

Re: [PATCH 04/11 v2] [POWERPC] Remove and replace uses of PPC_MEMSTART with memstart_addr

[Kumar Gala]

On Apr 1, 2008, at 5:50 AM, Paul Mackerras wrote:
> Kumar Gala writes:
>
>> For the sub-arches that support relocatable interrupt vectors (book- 
>> e) its
>> reasonable to have memory start at a non-zero physical address.   
>> For those
>> cases use the variable memstart_addr instead of the #define  
>> PPC_MEMSTART
>> since the only uses of PPC_MEMSTART are for initialization and in the
>> future we can set memstart_addr at runtime to have a relocatable  
>> kernel.
>
> In those cases, is it still true that the kernel sits at the start of
> the usable memory, or might there be usable memory before _stext?
> In other words, is PAGE_OFFSET == KERNELBASE still true or not?

Both cases are possible.

Here are the four cases I see:
* Normal kernel (kernel at physical 0)
* kexec kernel (kernel at physical 32M, but still has access to memory  
from physical 0 to 32M)

* cAMP kernel (kernel at non-zero offset) [eg. 256M offset]
* kexec cAMP kernel (kernel at 256M+32M, but has access to 256M + size)

- k

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Paul Mackerras]

Kumar Gala writes:

> Normally we assume kernel images will be loaded at offset 0. However
> there are situations, like when the kernel itself is running at a non-zero
> physical address, that we don't want to load it at 0.
> 
> Allow the wrapper to take an offset.  We use this when building u-boot images.

This makes it a bit harder to build a kernel and then wrap it later
on, since one would have to know what -m value to give.  Would it
possible for either the wrapper script or mkimage to peek in the ELF
header of the vmlinux to work out what physical address to use?

Paul.

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Kumar Gala]

On Apr 1, 2008, at 5:08 AM, Paul Mackerras wrote:
> Kumar Gala writes:
>
>> Normally we assume kernel images will be loaded at offset 0. However
>> there are situations, like when the kernel itself is running at a  
>> non-zero
>> physical address, that we don't want to load it at 0.
>>
>> Allow the wrapper to take an offset.  We use this when building u- 
>> boot images.
>
> This makes it a bit harder to build a kernel and then wrap it later
> on, since one would have to know what -m value to give.  Would it
> possible for either the wrapper script or mkimage to peek in the ELF
> header of the vmlinux to work out what physical address to use?

Hmm, need to think about that.  But my initial reaction is two fold.   
One I don't think this information would be around and two don't we  
already have this problem with a kdump kernel?

- k

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Paul Mackerras]

Kumar Gala writes:

> Hmm, need to think about that.  But my initial reaction is two fold.   
> One I don't think this information would be around and two don't we  
> already have this problem with a kdump kernel?

I'm just concerned that we have two things that have to match up - the
compiled-in physical address that the kernel assumes it is running at,
and the physical address where it is actually loaded.  While those two
things were both always 0 for embedded processors, there wasn't a
problem, but now we can have a situation where a kernel binary has to
be loaded at some nonzero address to work correctly, but there is no
way to work out what that address is for an existing vmlinux binary.
Or have I missed something?

For a kdump kernel, at least for 64-bit, the physical address has to
be 32MB.  There is no other choice, so there is no possibility of
confusion.

For 85xx, would it be possible to have the kernel figure out what
physical address it has been loaded at, and use that as the base
address, rather than having the base address set at compile time?
That would solve my objection since it would mean that there would no
longer be two things that had to be kept in sync.  You could pass any
value to wrapper/mkimage (subject to constraints such as it has to be
a multiple of 256M) and it would work.  That value could even come
from a config option in the case where wrapper is invoked as part of
the kernel build, but that config option shouldn't affect anything at
all in the vmlinux.

Paul.

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Kumar Gala]

On Apr 1, 2008, at 7:58 PM, Paul Mackerras wrote:
> Kumar Gala writes:
>
>> Hmm, need to think about that.  But my initial reaction is two fold.
>> One I don't think this information would be around and two don't we
>> already have this problem with a kdump kernel?
>
> I'm just concerned that we have two things that have to match up - the
> compiled-in physical address that the kernel assumes it is running at,
> and the physical address where it is actually loaded.  While those two
> things were both always 0 for embedded processors, there wasn't a
> problem, but now we can have a situation where a kernel binary has to
> be loaded at some nonzero address to work correctly, but there is no
> way to work out what that address is for an existing vmlinux binary.
> Or have I missed something?

Nope, that sums up the situation pretty well.

> For a kdump kernel, at least for 64-bit, the physical address has to
> be 32MB.  There is no other choice, so there is no possibility of
> confusion.

But how do you know a vmlinux image is for kdump or not?

> For 85xx, would it be possible to have the kernel figure out what
> physical address it has been loaded at, and use that as the base
> address, rather than having the base address set at compile time?

Yes, that is what CONFIG_RELOCATABLE is all about.

> That would solve my objection since it would mean that there would no
> longer be two things that had to be kept in sync.  You could pass any
> value to wrapper/mkimage (subject to constraints such as it has to be
> a multiple of 256M) and it would work.  That value could even come
> from a config option in the case where wrapper is invoked as part of
> the kernel build, but that config option shouldn't affect anything at
> all in the vmlinux.

Ok, but I still think the issues exists when we config PHYSICAL_START  
to non-zero and CONFIG_RELOCATABLE=n.  Ideally we get set the phys  
address the PHDR, but I'm not sure how to get the linker to do that.

- k

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Segher Boessenkool]

> Ideally we get set the phys address the PHDR, but I'm not sure how to 
> get the linker to do that.

I think you want to look at "AT" in the ld manual.


Segher

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Paul Mackerras]

Kumar Gala writes:

> > For a kdump kernel, at least for 64-bit, the physical address has to
> > be 32MB.  There is no other choice, so there is no possibility of
> > confusion.
> 
> But how do you know a vmlinux image is for kdump or not?

By looking at the ELF headers -- either the first PT_LOAD segment or
the .text section -- and seeing whether the start address is
0xc000000002000000 or not.

> > For 85xx, would it be possible to have the kernel figure out what
> > physical address it has been loaded at, and use that as the base
> > address, rather than having the base address set at compile time?
> 
> Yes, that is what CONFIG_RELOCATABLE is all about.

Is there any reason to have that as an option, rather than just always
doing that?

> > That would solve my objection since it would mean that there would no
> > longer be two things that had to be kept in sync.  You could pass any
> > value to wrapper/mkimage (subject to constraints such as it has to be
> > a multiple of 256M) and it would work.  That value could even come
> > from a config option in the case where wrapper is invoked as part of
> > the kernel build, but that config option shouldn't affect anything at
> > all in the vmlinux.
> 
> Ok, but I still think the issues exists when we config PHYSICAL_START  
> to non-zero and CONFIG_RELOCATABLE=n.  Ideally we get set the phys  
> address the PHDR, but I'm not sure how to get the linker to do that.

If we can do that then the wrapper script can dig it out and pass it
to mkimage, which would solve the problem.

Paul.

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Kumar Gala]

On Apr 3, 2008, at 1:34 AM, Paul Mackerras wrote:
> Kumar Gala writes:
>
>>> For a kdump kernel, at least for 64-bit, the physical address has to
>>> be 32MB.  There is no other choice, so there is no possibility of
>>> confusion.
>>
>> But how do you know a vmlinux image is for kdump or not?
>
> By looking at the ELF headers -- either the first PT_LOAD segment or
> the .text section -- and seeing whether the start address is
> 0xc000000002000000 or not.

fair point.

>>> For 85xx, would it be possible to have the kernel figure out what
>>> physical address it has been loaded at, and use that as the base
>>> address, rather than having the base address set at compile time?
>>
>> Yes, that is what CONFIG_RELOCATABLE is all about.
>
> Is there any reason to have that as an option, rather than just always
> doing that?

Concerned about runtime performance of __va() and __pa().

>>> That would solve my objection since it would mean that there would  
>>> no
>>> longer be two things that had to be kept in sync.  You could pass  
>>> any
>>> value to wrapper/mkimage (subject to constraints such as it has to  
>>> be
>>> a multiple of 256M) and it would work.  That value could even come
>>> from a config option in the case where wrapper is invoked as part of
>>> the kernel build, but that config option shouldn't affect anything  
>>> at
>>> all in the vmlinux.
>>
>> Ok, but I still think the issues exists when we config PHYSICAL_START
>> to non-zero and CONFIG_RELOCATABLE=n.  Ideally we get set the phys
>> address the PHDR, but I'm not sure how to get the linker to do that.
>
> If we can do that then the wrapper script can dig it out and pass it
> to mkimage, which would solve the problem.

Ok, so on Segher's recommendation I looked at 'AT' and posted a patch  
that uses that so now regardless of the kernel the PT_LOAD PHDR will  
have the physical address set properly.

So now we can look at the vmlinux and determine the physical offset.   
The question is how best to do that.  Here are the options I see:
* readelf, grep and parse output
* objdump grep and parse output
* simple C program that read's the elf and reports back

[looking for suggestion on what direction you want to take]

The other questions is if we'd ever have a vmlinux with more than one  
PT_LOAD PHDR.  If so which one do we use (the one with the lowest  
physical address)?

- k

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Paul Mackerras]

Kumar Gala writes:

> So now we can look at the vmlinux and determine the physical offset.   
> The question is how best to do that.  Here are the options I see:
> * readelf, grep and parse output
> * objdump grep and parse output
> * simple C program that read's the elf and reports back

Either readelf or objdump for now, and if that proves to be fragile we
can look at a C program.  You could do:

readelf -l $vmlinux | grep -m 1 LOAD | awk '{print $4}'

or

objdump -p $vmlinux | grep -m 1 LOAD | awk '{print $7}'

There's not a lot of difference.  Since the wrapper already uses
objdump, I think we should use objdump rather than making the wrapper
depend on an additional program (readelf).

> The other questions is if we'd ever have a vmlinux with more than one  
> PT_LOAD PHDR.  If so which one do we use (the one with the lowest  
> physical address)?

I think we would take the first one.

Paul.

Re: [PATCH 01/11] [POWERPC] bootwrapper: Allow specifying of image physical offset

[Kumar Gala]

On Apr 9, 2008, at 9:25 PM, Paul Mackerras wrote:
> Kumar Gala writes:
>
>> So now we can look at the vmlinux and determine the physical offset.
>> The question is how best to do that.  Here are the options I see:
>> * readelf, grep and parse output
>> * objdump grep and parse output
>> * simple C program that read's the elf and reports back
>
> Either readelf or objdump for now, and if that proves to be fragile we
> can look at a C program.  You could do:
>
> readelf -l $vmlinux | grep -m 1 LOAD | awk '{print $4}'
>
> or
>
> objdump -p $vmlinux | grep -m 1 LOAD | awk '{print $7}'
>
> There's not a lot of difference.  Since the wrapper already uses
> objdump, I think we should use objdump rather than making the wrapper
> depend on an additional program (readelf).
>
>> The other questions is if we'd ever have a vmlinux with more than one
>> PT_LOAD PHDR.  If so which one do we use (the one with the lowest
>> physical address)?
>
> I think we would take the first one.

Ok.  I've reworked this patch and sent in the new patch series.

- k