arm64: allow kernel Image to be loaded anywhere in physical memory

This relaxes the kernel Image placement requirements, so that it
may be placed at any 2 MB aligned offset in physical memory.

This is accomplished by ignoring PHYS_OFFSET when installing
memblocks, and accounting for the apparent virtual offset of
the kernel Image. As a result, virtual address references
below PAGE_OFFSET are correctly mapped onto physical references
into the kernel Image regardless of where it sits in memory.

Special care needs to be taken for dealing with memory limits passed
via mem=, since the generic implementation clips memory top down, which
may clip the kernel image itself if it is loaded high up in memory. To
deal with this case, we simply add back the memory covering the kernel
image, which may result in more memory to be retained than was passed
as a mem= parameter.

Since mem= should not be considered a production feature, a panic notifier
handler is installed that dumps the memory limit at panic time if one was
set.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>

9 files changed, 119 insertions, 39 deletions

diff --git a/Documentation/arm64/booting.txt b/Documentation/arm64/booting.txt
index 701d39d..56d6d8b 100644
--- a/Documentation/arm64/booting.txt
+++ b/Documentation/arm64/booting.txt
@@ -109,7 +109,13 @@ Header notes:
 			1 - 4K
 			2 - 16K
 			3 - 64K
-  Bits 3-63:	Reserved.
+  Bit 3:	Kernel physical placement
+			0 - 2MB aligned base should be as close as possible
+			    to the base of DRAM, since memory below it is not
+			    accessible via the linear mapping
+			1 - 2MB aligned base may be anywhere in physical
+			    memory
+  Bits 4-63:	Reserved.
 
 - When image_size is zero, a bootloader should attempt to keep as much
   memory as possible free for use by the kernel immediately after the
@@ -117,14 +123,14 @@ Header notes:
   depending on selected features, and is effectively unbound.
 
 The Image must be placed text_offset bytes from a 2MB aligned base
-address near the start of usable system RAM and called there. Memory
-below that base address is currently unusable by Linux, and therefore it
-is strongly recommended that this location is the start of system RAM.
-The region between the 2 MB aligned base address and the start of the
-image has no special significance to the kernel, and may be used for
-other purposes.
+address anywhere in usable system RAM and called there. The region
+between the 2 MB aligned base address and the start of the image has no
+special significance to the kernel, and may be used for other purposes.
 At least image_size bytes from the start of the image must be free for
 use by the kernel.
+NOTE: versions prior to v4.6 cannot make use of memory below the
+physical offset of the Image so it is recommended that the Image be
+placed as close as possible to the start of system RAM.
 
 Any memory described to the kernel (even that below the start of the
 image) which is not marked as reserved from the kernel (e.g., with a
diff --git a/arch/arm64/include/asm/boot.h b/arch/arm64/include/asm/boot.h
index 81151b6..ebf2481 100644
--- a/arch/arm64/include/asm/boot.h
+++ b/arch/arm64/include/asm/boot.h
@@ -11,4 +11,10 @@
 #define MIN_FDT_ALIGN		8
 #define MAX_FDT_SIZE		SZ_2M
 
+/*
+ * arm64 requires the kernel image to placed
+ * TEXT_OFFSET bytes beyond a 2 MB aligned base
+ */
+#define MIN_KIMG_ALIGN		SZ_2M
+
 #endif
diff --git a/arch/arm64/include/asm/kernel-pgtable.h b/arch/arm64/include/asm/kernel-pgtable.h
index a459714..5c6375d 100644
--- a/arch/arm64/include/asm/kernel-pgtable.h
+++ b/arch/arm64/include/asm/kernel-pgtable.h
@@ -79,5 +79,17 @@
 #define SWAPPER_MM_MMUFLAGS	(PTE_ATTRINDX(MT_NORMAL) | SWAPPER_PTE_FLAGS)
 #endif
 
+/*
+ * To make optimal use of block mappings when laying out the linear
+ * mapping, round down the base of physical memory to a size that can
+ * be mapped efficiently, i.e., either PUD_SIZE (4k granule) or PMD_SIZE
+ * (64k granule), or a multiple that can be mapped using contiguous bits
+ * in the page tables: 32 * PMD_SIZE (16k granule)
+ */
+#ifdef CONFIG_ARM64_64K_PAGES
+#define ARM64_MEMSTART_ALIGN	SZ_512M
+#else
+#define ARM64_MEMSTART_ALIGN	SZ_1G
+#endif
 
 #endif	/* __ASM_KERNEL_PGTABLE_H */
diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h
index 31b5600..054ac25 100644
--- a/arch/arm64/include/asm/kvm_asm.h
+++ b/arch/arm64/include/asm/kvm_asm.h
@@ -26,24 +26,9 @@
 #define KVM_ARM64_DEBUG_DIRTY_SHIFT	0
 #define KVM_ARM64_DEBUG_DIRTY		(1 << KVM_ARM64_DEBUG_DIRTY_SHIFT)
 
-#define kvm_ksym_ref(sym)		((void *)&sym + kvm_ksym_shift)
+#define kvm_ksym_ref(sym)		phys_to_virt((u64)&sym - kimage_voffset)
 
 #ifndef __ASSEMBLY__
-#if __GNUC__ > 4
-#define kvm_ksym_shift			(PAGE_OFFSET - KIMAGE_VADDR)
-#else
-/*
- * GCC versions 4.9 and older will fold the constant below into the addend of
- * the reference to 'sym' above if kvm_ksym_shift is declared static or if the
- * constant is used directly. However, since we use the small code model for
- * the core kernel, the reference to 'sym' will be emitted as a adrp/add pair,
- * with a +/- 4 GB range, resulting in linker relocation errors if the shift
- * is sufficiently large. So prevent the compiler from folding the shift into
- * the addend, by making the shift a variable with external linkage.
- */
-__weak u64 kvm_ksym_shift = PAGE_OFFSET - KIMAGE_VADDR;
-#endif
-
 struct kvm;
 struct kvm_vcpu;
 
diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memory.h
index 18b7e77..3239e4d 100644
--- a/arch/arm64/include/asm/memory.h
+++ b/arch/arm64/include/asm/memory.h
@@ -24,6 +24,7 @@
 #include <linux/compiler.h>
 #include <linux/const.h>
 #include <linux/types.h>
+#include <asm/bug.h>
 #include <asm/sizes.h>
 
 /*
@@ -88,10 +89,10 @@
 #define __virt_to_phys(x) ({						\
 	phys_addr_t __x = (phys_addr_t)(x);				\
 	__x >= PAGE_OFFSET ? (__x - PAGE_OFFSET + PHYS_OFFSET) :	\
-			     (__x - KIMAGE_VADDR + PHYS_OFFSET); })
+			     (__x - kimage_voffset); })
 
 #define __phys_to_virt(x)	((unsigned long)((x) - PHYS_OFFSET + PAGE_OFFSET))
-#define __phys_to_kimg(x)	((unsigned long)((x) - PHYS_OFFSET + KIMAGE_VADDR))
+#define __phys_to_kimg(x)	((unsigned long)((x) + kimage_voffset))
 
 /*
  * Convert a page to/from a physical address
@@ -133,15 +134,16 @@
 
 extern phys_addr_t		memstart_addr;
 /* PHYS_OFFSET - the physical address of the start of memory. */
-#define PHYS_OFFSET		({ memstart_addr; })
+#define PHYS_OFFSET		({ BUG_ON(memstart_addr & 1); memstart_addr; })
+
+/* the offset between the kernel virtual and physical mappings */
+extern u64			kimage_voffset;
 
 /*
- * The maximum physical address that the linear direct mapping
- * of system RAM can cover. (PAGE_OFFSET can be interpreted as
- * a 2's complement signed quantity and negated to derive the
- * maximum size of the linear mapping.)
+ * Allow all memory at the discovery stage. We will clip it later.
  */
-#define MAX_MEMBLOCK_ADDR	({ memstart_addr - PAGE_OFFSET - 1; })
+#define MIN_MEMBLOCK_ADDR	0
+#define MAX_MEMBLOCK_ADDR	U64_MAX
 
 /*
  * PFNs are used to describe any physical page; this means
diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S
index 04d38a0..05b9828 100644
--- a/arch/arm64/kernel/head.S
+++ b/arch/arm64/kernel/head.S
@@ -428,7 +428,11 @@ __mmap_switched:
 	and	x4, x4, #~(THREAD_SIZE - 1)
 	msr	sp_el0, x4			// Save thread_info
 	str_l	x21, __fdt_pointer, x5		// Save FDT pointer
-	str_l	x24, memstart_addr, x6		// Save PHYS_OFFSET
+
+	ldr	x4, =KIMAGE_VADDR		// Save the offset between
+	sub	x4, x4, x24			// the kernel virtual and
+	str_l	x4, kimage_voffset, x5		// physical mappings
+
 	mov	x29, #0
 #ifdef CONFIG_KASAN
 	bl	kasan_early_init
diff --git a/arch/arm64/kernel/image.h b/arch/arm64/kernel/image.h
index 999633b..c9c62ca 100644
--- a/arch/arm64/kernel/image.h
+++ b/arch/arm64/kernel/image.h
@@ -42,15 +42,18 @@
 #endif
 
 #ifdef CONFIG_CPU_BIG_ENDIAN
-#define __HEAD_FLAG_BE	1
+#define __HEAD_FLAG_BE		1
 #else
-#define __HEAD_FLAG_BE	0
+#define __HEAD_FLAG_BE		0
 #endif
 
-#define __HEAD_FLAG_PAGE_SIZE ((PAGE_SHIFT - 10) / 2)
+#define __HEAD_FLAG_PAGE_SIZE	((PAGE_SHIFT - 10) / 2)
 
-#define __HEAD_FLAGS	((__HEAD_FLAG_BE << 0) |	\
-			 (__HEAD_FLAG_PAGE_SIZE << 1))
+#define __HEAD_FLAG_PHYS_BASE	1
+
+#define __HEAD_FLAGS		((__HEAD_FLAG_BE << 0) |	\
+				 (__HEAD_FLAG_PAGE_SIZE << 1) |	\
+				 (__HEAD_FLAG_PHYS_BASE << 3))
 
 /*
  * These will output as part of the Image header, which should be little-endian
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index 52d1fc4..c0ea54b 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -35,8 +35,10 @@
 #include <linux/efi.h>
 #include <linux/swiotlb.h>
 
+#include <asm/boot.h>
 #include <asm/fixmap.h>
 #include <asm/kasan.h>
+#include <asm/kernel-pgtable.h>
 #include <asm/memory.h>
 #include <asm/sections.h>
 #include <asm/setup.h>
@@ -46,7 +48,13 @@
 
 #include "mm.h"
 
-phys_addr_t memstart_addr __read_mostly = 0;
+/*
+ * We need to be able to catch inadvertent references to memstart_addr
+ * that occur (potentially in generic code) before arm64_memblock_init()
+ * executes, which assigns it its actual value. So use a default value
+ * that cannot be mistaken for a real physical address.
+ */
+phys_addr_t memstart_addr __read_mostly = ~0ULL;
 phys_addr_t arm64_dma_phys_limit __read_mostly;
 
 #ifdef CONFIG_BLK_DEV_INITRD
@@ -160,7 +168,33 @@ early_param("mem", early_mem);
 
 void __init arm64_memblock_init(void)
 {
-	memblock_enforce_memory_limit(memory_limit);
+	const s64 linear_region_size = -(s64)PAGE_OFFSET;
+
+	/*
+	 * Select a suitable value for the base of physical memory.
+	 */
+	memstart_addr = round_down(memblock_start_of_DRAM(),
+				   ARM64_MEMSTART_ALIGN);
+
+	/*
+	 * Remove the memory that we will not be able to cover with the
+	 * linear mapping. Take care not to clip the kernel which may be
+	 * high in memory.
+	 */
+	memblock_remove(max(memstart_addr + linear_region_size, __pa(_end)),
+			ULLONG_MAX);
+	if (memblock_end_of_DRAM() > linear_region_size)
+		memblock_remove(0, memblock_end_of_DRAM() - linear_region_size);
+
+	/*
+	 * Apply the memory limit if it was set. Since the kernel may be loaded
+	 * high up in memory, add back the kernel region that must be accessible
+	 * via the linear mapping.
+	 */
+	if (memory_limit != (phys_addr_t)ULLONG_MAX) {
+		memblock_enforce_memory_limit(memory_limit);
+		memblock_add(__pa(_text), (u64)(_end - _text));
+	}
 
 	/*
 	 * Register the kernel text, kernel data, initrd, and initial
@@ -386,3 +420,28 @@ static int __init keepinitrd_setup(char *__unused)
 
 __setup("keepinitrd", keepinitrd_setup);
 #endif
+
+/*
+ * Dump out memory limit information on panic.
+ */
+static int dump_mem_limit(struct notifier_block *self, unsigned long v, void *p)
+{
+	if (memory_limit != (phys_addr_t)ULLONG_MAX) {
+		pr_emerg("Memory Limit: %llu MB\n", memory_limit >> 20);
+	} else {
+		pr_emerg("Memory Limit: none\n");
+	}
+	return 0;
+}
+
+static struct notifier_block mem_limit_notifier = {
+	.notifier_call = dump_mem_limit,
+};
+
+static int __init register_mem_limit_dumper(void)
+{
+	atomic_notifier_chain_register(&panic_notifier_list,
+				       &mem_limit_notifier);
+	return 0;
+}
+__initcall(register_mem_limit_dumper);
diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
index 1d9aea4..32c4a37 100644
--- a/arch/arm64/mm/mmu.c
+++ b/arch/arm64/mm/mmu.c
@@ -46,6 +46,9 @@
 
 u64 idmap_t0sz = TCR_T0SZ(VA_BITS);
 
+u64 kimage_voffset __read_mostly;
+EXPORT_SYMBOL(kimage_voffset);
+
 /*
  * Empty_zero_page is a special page that is used for zero-initialized data
  * and COW.