linux-brain/arch/arm/include/asm/memory.h

378 lines
10 KiB
C
Raw Normal View History

/* SPDX-License-Identifier: GPL-2.0-only */
/*
* arch/arm/include/asm/memory.h
*
* Copyright (C) 2000-2002 Russell King
* modification for nommu, Hyok S. Choi, 2004
*
* Note: this file should not be included by non-asm/.h files
*/
#ifndef __ASM_ARM_MEMORY_H
#define __ASM_ARM_MEMORY_H
#include <linux/compiler.h>
#include <linux/const.h>
#include <linux/types.h>
#include <linux/sizes.h>
#ifdef CONFIG_NEED_MACH_MEMORY_H
#include <mach/memory.h>
#endif
/* PAGE_OFFSET - the virtual address of the start of the kernel image */
#define PAGE_OFFSET UL(CONFIG_PAGE_OFFSET)
#ifdef CONFIG_MMU
/*
* TASK_SIZE - the maximum size of a user space task.
* TASK_UNMAPPED_BASE - the lower boundary of the mmap VM area
*/
#define TASK_SIZE (UL(CONFIG_PAGE_OFFSET) - UL(SZ_16M))
#define TASK_UNMAPPED_BASE ALIGN(TASK_SIZE / 3, SZ_16M)
/*
* The maximum size of a 26-bit user space task.
*/
#define TASK_SIZE_26 (UL(1) << 26)
/*
* The module space lives between the addresses given by TASK_SIZE
* and PAGE_OFFSET - it must be within 32MB of the kernel text.
*/
#ifndef CONFIG_THUMB2_KERNEL
#define MODULES_VADDR (PAGE_OFFSET - SZ_16M)
#else
/* smaller range for Thumb-2 symbols relocation (2^24)*/
#define MODULES_VADDR (PAGE_OFFSET - SZ_8M)
#endif
#if TASK_SIZE > MODULES_VADDR
#error Top of user space clashes with start of module space
#endif
/*
* The highmem pkmap virtual space shares the end of the module area.
*/
#ifdef CONFIG_HIGHMEM
#define MODULES_END (PAGE_OFFSET - PMD_SIZE)
#else
#define MODULES_END (PAGE_OFFSET)
#endif
/*
* The XIP kernel gets mapped at the bottom of the module vm area.
* Since we use sections to map it, this macro replaces the physical address
* with its virtual address while keeping offset from the base section.
*/
#define XIP_VIRT_ADDR(physaddr) (MODULES_VADDR + ((physaddr) & 0x000fffff))
#define FDT_FIXED_BASE UL(0xff800000)
#define FDT_FIXED_SIZE (2 * SECTION_SIZE)
#define FDT_VIRT_BASE(physbase) ((void *)(FDT_FIXED_BASE | (physbase) % SECTION_SIZE))
#if !defined(CONFIG_SMP) && !defined(CONFIG_ARM_LPAE)
/*
* Allow 16MB-aligned ioremap pages
*/
#define IOREMAP_MAX_ORDER 24
#endif
#define VECTORS_BASE UL(0xffff0000)
#else /* CONFIG_MMU */
#ifndef __ASSEMBLY__
extern unsigned long setup_vectors_base(void);
extern unsigned long vectors_base;
#define VECTORS_BASE vectors_base
#endif
/*
* The limitation of user task size can grow up to the end of free ram region.
* It is difficult to define and perhaps will never meet the original meaning
* of this define that was meant to.
* Fortunately, there is no reference for this in noMMU mode, for now.
*/
#define TASK_SIZE UL(0xffffffff)
#ifndef TASK_UNMAPPED_BASE
#define TASK_UNMAPPED_BASE UL(0x00000000)
#endif
#ifndef END_MEM
#define END_MEM (UL(CONFIG_DRAM_BASE) + CONFIG_DRAM_SIZE)
#endif
/*
* The module can be at any place in ram in nommu mode.
*/
#define MODULES_END (END_MEM)
#define MODULES_VADDR PAGE_OFFSET
#define XIP_VIRT_ADDR(physaddr) (physaddr)
#define FDT_VIRT_BASE(physbase) ((void *)(physbase))
#endif /* !CONFIG_MMU */
#ifdef CONFIG_XIP_KERNEL
#define KERNEL_START _sdata
#else
#define KERNEL_START _stext
#endif
#define KERNEL_END _end
/*
* We fix the TCM memories max 32 KiB ITCM resp DTCM at these
* locations
*/
#ifdef CONFIG_HAVE_TCM
#define ITCM_OFFSET UL(0xfffe0000)
#define DTCM_OFFSET UL(0xfffe8000)
#endif
/*
* Convert a page to/from a physical address
*/
#define page_to_phys(page) (__pfn_to_phys(page_to_pfn(page)))
#define phys_to_page(phys) (pfn_to_page(__phys_to_pfn(phys)))
/*
* PLAT_PHYS_OFFSET is the offset (from zero) of the start of physical
* memory. This is used for XIP and NoMMU kernels, and on platforms that don't
* have CONFIG_ARM_PATCH_PHYS_VIRT. Assembly code must always use
* PLAT_PHYS_OFFSET and not PHYS_OFFSET.
*/
#define PLAT_PHYS_OFFSET UL(CONFIG_PHYS_OFFSET)
#ifdef CONFIG_XIP_KERNEL
/*
* When referencing data in RAM from the XIP region in a relative manner
* with the MMU off, we need the relative offset between the two physical
* addresses. The macro below achieves this, which is:
* __pa(v_data) - __xip_pa(v_text)
*/
#define PHYS_RELATIVE(v_data, v_text) \
(((v_data) - PAGE_OFFSET + PLAT_PHYS_OFFSET) - \
((v_text) - XIP_VIRT_ADDR(CONFIG_XIP_PHYS_ADDR) + \
CONFIG_XIP_PHYS_ADDR))
#else
#define PHYS_RELATIVE(v_data, v_text) ((v_data) - (v_text))
#endif
#ifndef __ASSEMBLY__
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
/*
* Physical vs virtual RAM address space conversion. These are
* private definitions which should NOT be used outside memory.h
* files. Use virt_to_phys/phys_to_virt/__pa/__va instead.
ARM: Better virt_to_page() handling virt_to_page() is incredibly inefficient when virt-to-phys patching is enabled. This is because we end up with this calculation: page = &mem_map[asm virt_to_phys(addr) >> 12 - __pv_phys_offset >> 12] in assembly. The asm virt_to_phys() is equivalent this this operation: addr - PAGE_OFFSET + __pv_phys_offset and we can see that because this is assembly, the compiler has no chance to optimise some of that away. This should reduce down to: page = &mem_map[(addr - PAGE_OFFSET) >> 12] for the common cases. Permit the compiler to make this optimisation by giving it more of the information it needs - do this by providing a virt_to_pfn() macro. Another issue which makes this more complex is that __pv_phys_offset is a 64-bit type on all platforms. This is needlessly wasteful - if we store the physical offset as a PFN, we can save a lot of work having to deal with 64-bit values, which sometimes ends up producing incredibly horrid code: a4c: e3009000 movw r9, #0 a4c: R_ARM_MOVW_ABS_NC __pv_phys_offset a50: e3409000 movt r9, #0 ; r9 = &__pv_phys_offset a50: R_ARM_MOVT_ABS __pv_phys_offset a54: e3002000 movw r2, #0 a54: R_ARM_MOVW_ABS_NC __pv_phys_offset a58: e3402000 movt r2, #0 ; r2 = &__pv_phys_offset a58: R_ARM_MOVT_ABS __pv_phys_offset a5c: e5999004 ldr r9, [r9, #4] ; r9 = high word of __pv_phys_offset a60: e3001000 movw r1, #0 a60: R_ARM_MOVW_ABS_NC mem_map a64: e592c000 ldr ip, [r2] ; ip = low word of __pv_phys_offset Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-03-26 04:45:31 +09:00
*
* PFNs are used to describe any physical page; this means
* PFN 0 == physical address 0.
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
*/
ARM: Better virt_to_page() handling virt_to_page() is incredibly inefficient when virt-to-phys patching is enabled. This is because we end up with this calculation: page = &mem_map[asm virt_to_phys(addr) >> 12 - __pv_phys_offset >> 12] in assembly. The asm virt_to_phys() is equivalent this this operation: addr - PAGE_OFFSET + __pv_phys_offset and we can see that because this is assembly, the compiler has no chance to optimise some of that away. This should reduce down to: page = &mem_map[(addr - PAGE_OFFSET) >> 12] for the common cases. Permit the compiler to make this optimisation by giving it more of the information it needs - do this by providing a virt_to_pfn() macro. Another issue which makes this more complex is that __pv_phys_offset is a 64-bit type on all platforms. This is needlessly wasteful - if we store the physical offset as a PFN, we can save a lot of work having to deal with 64-bit values, which sometimes ends up producing incredibly horrid code: a4c: e3009000 movw r9, #0 a4c: R_ARM_MOVW_ABS_NC __pv_phys_offset a50: e3409000 movt r9, #0 ; r9 = &__pv_phys_offset a50: R_ARM_MOVT_ABS __pv_phys_offset a54: e3002000 movw r2, #0 a54: R_ARM_MOVW_ABS_NC __pv_phys_offset a58: e3402000 movt r2, #0 ; r2 = &__pv_phys_offset a58: R_ARM_MOVT_ABS __pv_phys_offset a5c: e5999004 ldr r9, [r9, #4] ; r9 = high word of __pv_phys_offset a60: e3001000 movw r1, #0 a60: R_ARM_MOVW_ABS_NC mem_map a64: e592c000 ldr ip, [r2] ; ip = low word of __pv_phys_offset Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-03-26 04:45:31 +09:00
#if defined(CONFIG_ARM_PATCH_PHYS_VIRT)
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
/*
* Constants used to force the right instruction encodings and shifts
* so that all we need to do is modify the 8-bit constant field.
*/
#define __PV_BITS_31_24 0x81000000
ARM: mm: Correct virt_to_phys patching for 64 bit physical addresses The current phys_to_virt patching mechanism works only for 32 bit physical addresses and this patch extends the idea for 64bit physical addresses. The 64bit v2p patching mechanism patches the higher 8 bits of physical address with a constant using 'mov' instruction and lower 32bits are patched using 'add'. While this is correct, in those platforms where the lowmem addressable physical memory spawns across 4GB boundary, a carry bit can be produced as a result of addition of lower 32bits. This has to be taken in to account and added in to the upper. The patched __pv_offset and va are added in lower 32bits, where __pv_offset can be in two's complement form when PA_START < VA_START and that can result in a false carry bit. e.g 1) PA = 0x80000000; VA = 0xC0000000 __pv_offset = PA - VA = 0xC0000000 (2's complement) 2) PA = 0x2 80000000; VA = 0xC000000 __pv_offset = PA - VA = 0x1 C0000000 So adding __pv_offset + VA should never result in a true overflow for (1). So in order to differentiate between a true carry, a __pv_offset is extended to 64bit and the upper 32bits will have 0xffffffff if __pv_offset is 2's complement. So 'mvn #0' is inserted instead of 'mov' while patching for the same reason. Since mov, add, sub instruction are to patched with different constants inside the same stub, the rotation field of the opcode is using to differentiate between them. So the above examples for v2p translation becomes for VA=0xC0000000, 1) PA[63:32] = 0xffffffff PA[31:0] = VA + 0xC0000000 --> results in a carry PA[63:32] = PA[63:32] + carry PA[63:0] = 0x0 80000000 2) PA[63:32] = 0x1 PA[31:0] = VA + 0xC0000000 --> results in a carry PA[63:32] = PA[63:32] + carry PA[63:0] = 0x2 80000000 The above ideas were suggested by Nicolas Pitre <nico@linaro.org> as part of the review of first and second versions of the subject patch. There is no corresponding change on the phys_to_virt() side, because computations on the upper 32-bits would be discarded anyway. Cc: Russell King <linux@arm.linux.org.uk> Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Sricharan R <r.sricharan@ti.com> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
2013-07-29 23:56:22 +09:00
#define __PV_BITS_7_0 0x81
ARM: Better virt_to_page() handling virt_to_page() is incredibly inefficient when virt-to-phys patching is enabled. This is because we end up with this calculation: page = &mem_map[asm virt_to_phys(addr) >> 12 - __pv_phys_offset >> 12] in assembly. The asm virt_to_phys() is equivalent this this operation: addr - PAGE_OFFSET + __pv_phys_offset and we can see that because this is assembly, the compiler has no chance to optimise some of that away. This should reduce down to: page = &mem_map[(addr - PAGE_OFFSET) >> 12] for the common cases. Permit the compiler to make this optimisation by giving it more of the information it needs - do this by providing a virt_to_pfn() macro. Another issue which makes this more complex is that __pv_phys_offset is a 64-bit type on all platforms. This is needlessly wasteful - if we store the physical offset as a PFN, we can save a lot of work having to deal with 64-bit values, which sometimes ends up producing incredibly horrid code: a4c: e3009000 movw r9, #0 a4c: R_ARM_MOVW_ABS_NC __pv_phys_offset a50: e3409000 movt r9, #0 ; r9 = &__pv_phys_offset a50: R_ARM_MOVT_ABS __pv_phys_offset a54: e3002000 movw r2, #0 a54: R_ARM_MOVW_ABS_NC __pv_phys_offset a58: e3402000 movt r2, #0 ; r2 = &__pv_phys_offset a58: R_ARM_MOVT_ABS __pv_phys_offset a5c: e5999004 ldr r9, [r9, #4] ; r9 = high word of __pv_phys_offset a60: e3001000 movw r1, #0 a60: R_ARM_MOVW_ABS_NC mem_map a64: e592c000 ldr ip, [r2] ; ip = low word of __pv_phys_offset Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-03-26 04:45:31 +09:00
extern unsigned long __pv_phys_pfn_offset;
ARM: mm: Correct virt_to_phys patching for 64 bit physical addresses The current phys_to_virt patching mechanism works only for 32 bit physical addresses and this patch extends the idea for 64bit physical addresses. The 64bit v2p patching mechanism patches the higher 8 bits of physical address with a constant using 'mov' instruction and lower 32bits are patched using 'add'. While this is correct, in those platforms where the lowmem addressable physical memory spawns across 4GB boundary, a carry bit can be produced as a result of addition of lower 32bits. This has to be taken in to account and added in to the upper. The patched __pv_offset and va are added in lower 32bits, where __pv_offset can be in two's complement form when PA_START < VA_START and that can result in a false carry bit. e.g 1) PA = 0x80000000; VA = 0xC0000000 __pv_offset = PA - VA = 0xC0000000 (2's complement) 2) PA = 0x2 80000000; VA = 0xC000000 __pv_offset = PA - VA = 0x1 C0000000 So adding __pv_offset + VA should never result in a true overflow for (1). So in order to differentiate between a true carry, a __pv_offset is extended to 64bit and the upper 32bits will have 0xffffffff if __pv_offset is 2's complement. So 'mvn #0' is inserted instead of 'mov' while patching for the same reason. Since mov, add, sub instruction are to patched with different constants inside the same stub, the rotation field of the opcode is using to differentiate between them. So the above examples for v2p translation becomes for VA=0xC0000000, 1) PA[63:32] = 0xffffffff PA[31:0] = VA + 0xC0000000 --> results in a carry PA[63:32] = PA[63:32] + carry PA[63:0] = 0x0 80000000 2) PA[63:32] = 0x1 PA[31:0] = VA + 0xC0000000 --> results in a carry PA[63:32] = PA[63:32] + carry PA[63:0] = 0x2 80000000 The above ideas were suggested by Nicolas Pitre <nico@linaro.org> as part of the review of first and second versions of the subject patch. There is no corresponding change on the phys_to_virt() side, because computations on the upper 32-bits would be discarded anyway. Cc: Russell King <linux@arm.linux.org.uk> Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Sricharan R <r.sricharan@ti.com> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
2013-07-29 23:56:22 +09:00
extern u64 __pv_offset;
extern void fixup_pv_table(const void *, unsigned long);
extern const void *__pv_table_begin, *__pv_table_end;
ARM: Better virt_to_page() handling virt_to_page() is incredibly inefficient when virt-to-phys patching is enabled. This is because we end up with this calculation: page = &mem_map[asm virt_to_phys(addr) >> 12 - __pv_phys_offset >> 12] in assembly. The asm virt_to_phys() is equivalent this this operation: addr - PAGE_OFFSET + __pv_phys_offset and we can see that because this is assembly, the compiler has no chance to optimise some of that away. This should reduce down to: page = &mem_map[(addr - PAGE_OFFSET) >> 12] for the common cases. Permit the compiler to make this optimisation by giving it more of the information it needs - do this by providing a virt_to_pfn() macro. Another issue which makes this more complex is that __pv_phys_offset is a 64-bit type on all platforms. This is needlessly wasteful - if we store the physical offset as a PFN, we can save a lot of work having to deal with 64-bit values, which sometimes ends up producing incredibly horrid code: a4c: e3009000 movw r9, #0 a4c: R_ARM_MOVW_ABS_NC __pv_phys_offset a50: e3409000 movt r9, #0 ; r9 = &__pv_phys_offset a50: R_ARM_MOVT_ABS __pv_phys_offset a54: e3002000 movw r2, #0 a54: R_ARM_MOVW_ABS_NC __pv_phys_offset a58: e3402000 movt r2, #0 ; r2 = &__pv_phys_offset a58: R_ARM_MOVT_ABS __pv_phys_offset a5c: e5999004 ldr r9, [r9, #4] ; r9 = high word of __pv_phys_offset a60: e3001000 movw r1, #0 a60: R_ARM_MOVW_ABS_NC mem_map a64: e592c000 ldr ip, [r2] ; ip = low word of __pv_phys_offset Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-03-26 04:45:31 +09:00
#define PHYS_OFFSET ((phys_addr_t)__pv_phys_pfn_offset << PAGE_SHIFT)
#define PHYS_PFN_OFFSET (__pv_phys_pfn_offset)
#define __pv_stub(from,to,instr,type) \
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
__asm__("@ __pv_stub\n" \
"1: " instr " %0, %1, %2\n" \
" .pushsection .pv_table,\"a\"\n" \
" .long 1b\n" \
" .popsection\n" \
: "=r" (to) \
: "r" (from), "I" (type))
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
ARM: mm: Correct virt_to_phys patching for 64 bit physical addresses The current phys_to_virt patching mechanism works only for 32 bit physical addresses and this patch extends the idea for 64bit physical addresses. The 64bit v2p patching mechanism patches the higher 8 bits of physical address with a constant using 'mov' instruction and lower 32bits are patched using 'add'. While this is correct, in those platforms where the lowmem addressable physical memory spawns across 4GB boundary, a carry bit can be produced as a result of addition of lower 32bits. This has to be taken in to account and added in to the upper. The patched __pv_offset and va are added in lower 32bits, where __pv_offset can be in two's complement form when PA_START < VA_START and that can result in a false carry bit. e.g 1) PA = 0x80000000; VA = 0xC0000000 __pv_offset = PA - VA = 0xC0000000 (2's complement) 2) PA = 0x2 80000000; VA = 0xC000000 __pv_offset = PA - VA = 0x1 C0000000 So adding __pv_offset + VA should never result in a true overflow for (1). So in order to differentiate between a true carry, a __pv_offset is extended to 64bit and the upper 32bits will have 0xffffffff if __pv_offset is 2's complement. So 'mvn #0' is inserted instead of 'mov' while patching for the same reason. Since mov, add, sub instruction are to patched with different constants inside the same stub, the rotation field of the opcode is using to differentiate between them. So the above examples for v2p translation becomes for VA=0xC0000000, 1) PA[63:32] = 0xffffffff PA[31:0] = VA + 0xC0000000 --> results in a carry PA[63:32] = PA[63:32] + carry PA[63:0] = 0x0 80000000 2) PA[63:32] = 0x1 PA[31:0] = VA + 0xC0000000 --> results in a carry PA[63:32] = PA[63:32] + carry PA[63:0] = 0x2 80000000 The above ideas were suggested by Nicolas Pitre <nico@linaro.org> as part of the review of first and second versions of the subject patch. There is no corresponding change on the phys_to_virt() side, because computations on the upper 32-bits would be discarded anyway. Cc: Russell King <linux@arm.linux.org.uk> Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Sricharan R <r.sricharan@ti.com> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
2013-07-29 23:56:22 +09:00
#define __pv_stub_mov_hi(t) \
__asm__ volatile("@ __pv_stub_mov\n" \
"1: mov %R0, %1\n" \
" .pushsection .pv_table,\"a\"\n" \
" .long 1b\n" \
" .popsection\n" \
: "=r" (t) \
: "I" (__PV_BITS_7_0))
#define __pv_add_carry_stub(x, y) \
__asm__ volatile("@ __pv_add_carry_stub\n" \
"1: adds %Q0, %1, %2\n" \
" adc %R0, %R0, #0\n" \
" .pushsection .pv_table,\"a\"\n" \
" .long 1b\n" \
" .popsection\n" \
: "+r" (y) \
: "r" (x), "I" (__PV_BITS_31_24) \
: "cc")
static inline phys_addr_t __virt_to_phys_nodebug(unsigned long x)
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
{
ARM: mm: Correct virt_to_phys patching for 64 bit physical addresses The current phys_to_virt patching mechanism works only for 32 bit physical addresses and this patch extends the idea for 64bit physical addresses. The 64bit v2p patching mechanism patches the higher 8 bits of physical address with a constant using 'mov' instruction and lower 32bits are patched using 'add'. While this is correct, in those platforms where the lowmem addressable physical memory spawns across 4GB boundary, a carry bit can be produced as a result of addition of lower 32bits. This has to be taken in to account and added in to the upper. The patched __pv_offset and va are added in lower 32bits, where __pv_offset can be in two's complement form when PA_START < VA_START and that can result in a false carry bit. e.g 1) PA = 0x80000000; VA = 0xC0000000 __pv_offset = PA - VA = 0xC0000000 (2's complement) 2) PA = 0x2 80000000; VA = 0xC000000 __pv_offset = PA - VA = 0x1 C0000000 So adding __pv_offset + VA should never result in a true overflow for (1). So in order to differentiate between a true carry, a __pv_offset is extended to 64bit and the upper 32bits will have 0xffffffff if __pv_offset is 2's complement. So 'mvn #0' is inserted instead of 'mov' while patching for the same reason. Since mov, add, sub instruction are to patched with different constants inside the same stub, the rotation field of the opcode is using to differentiate between them. So the above examples for v2p translation becomes for VA=0xC0000000, 1) PA[63:32] = 0xffffffff PA[31:0] = VA + 0xC0000000 --> results in a carry PA[63:32] = PA[63:32] + carry PA[63:0] = 0x0 80000000 2) PA[63:32] = 0x1 PA[31:0] = VA + 0xC0000000 --> results in a carry PA[63:32] = PA[63:32] + carry PA[63:0] = 0x2 80000000 The above ideas were suggested by Nicolas Pitre <nico@linaro.org> as part of the review of first and second versions of the subject patch. There is no corresponding change on the phys_to_virt() side, because computations on the upper 32-bits would be discarded anyway. Cc: Russell King <linux@arm.linux.org.uk> Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Sricharan R <r.sricharan@ti.com> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
2013-07-29 23:56:22 +09:00
phys_addr_t t;
if (sizeof(phys_addr_t) == 4) {
__pv_stub(x, t, "add", __PV_BITS_31_24);
} else {
__pv_stub_mov_hi(t);
__pv_add_carry_stub(x, t);
}
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
return t;
}
static inline unsigned long __phys_to_virt(phys_addr_t x)
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
{
unsigned long t;
/*
* 'unsigned long' cast discard upper word when
* phys_addr_t is 64 bit, and makes sure that inline
* assembler expression receives 32 bit argument
* in place where 'r' 32 bit operand is expected.
*/
__pv_stub((unsigned long) x, t, "sub", __PV_BITS_31_24);
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
return t;
}
ARM: P2V: introduce phys_to_virt/virt_to_phys runtime patching This idea came from Nicolas, Eric Miao produced an initial version, which was then rewritten into this. Patch the physical to virtual translations at runtime. As we modify the code, this makes it incompatible with XIP kernels, but allows us to achieve this with minimal loss of performance. As many translations are of the form: physical = virtual + (PHYS_OFFSET - PAGE_OFFSET) virtual = physical - (PHYS_OFFSET - PAGE_OFFSET) we generate an 'add' instruction for __virt_to_phys(), and a 'sub' instruction for __phys_to_virt(). We calculate at run time (PHYS_OFFSET - PAGE_OFFSET) by comparing the address prior to MMU initialization with where it should be once the MMU has been initialized, and place this constant into the above add/sub instructions. Once we have (PHYS_OFFSET - PAGE_OFFSET), we can calculate the real PHYS_OFFSET as PAGE_OFFSET is a build-time constant, and save this for the C-mode PHYS_OFFSET variable definition to use. At present, we are unable to support Realview with Sparsemem enabled as this uses a complex mapping function, and MSM as this requires a constant which will not fit in our math instruction. Add a module version magic string for this feature to prevent incompatible modules being loaded. Tested-by: Tony Lindgren <tony@atomide.com> Reviewed-by: Nicolas Pitre <nicolas.pitre@linaro.org> Tested-by: Nicolas Pitre <nicolas.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2011-01-05 04:09:43 +09:00
#else
#define PHYS_OFFSET PLAT_PHYS_OFFSET
ARM: Better virt_to_page() handling virt_to_page() is incredibly inefficient when virt-to-phys patching is enabled. This is because we end up with this calculation: page = &mem_map[asm virt_to_phys(addr) >> 12 - __pv_phys_offset >> 12] in assembly. The asm virt_to_phys() is equivalent this this operation: addr - PAGE_OFFSET + __pv_phys_offset and we can see that because this is assembly, the compiler has no chance to optimise some of that away. This should reduce down to: page = &mem_map[(addr - PAGE_OFFSET) >> 12] for the common cases. Permit the compiler to make this optimisation by giving it more of the information it needs - do this by providing a virt_to_pfn() macro. Another issue which makes this more complex is that __pv_phys_offset is a 64-bit type on all platforms. This is needlessly wasteful - if we store the physical offset as a PFN, we can save a lot of work having to deal with 64-bit values, which sometimes ends up producing incredibly horrid code: a4c: e3009000 movw r9, #0 a4c: R_ARM_MOVW_ABS_NC __pv_phys_offset a50: e3409000 movt r9, #0 ; r9 = &__pv_phys_offset a50: R_ARM_MOVT_ABS __pv_phys_offset a54: e3002000 movw r2, #0 a54: R_ARM_MOVW_ABS_NC __pv_phys_offset a58: e3402000 movt r2, #0 ; r2 = &__pv_phys_offset a58: R_ARM_MOVT_ABS __pv_phys_offset a5c: e5999004 ldr r9, [r9, #4] ; r9 = high word of __pv_phys_offset a60: e3001000 movw r1, #0 a60: R_ARM_MOVW_ABS_NC mem_map a64: e592c000 ldr ip, [r2] ; ip = low word of __pv_phys_offset Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-03-26 04:45:31 +09:00
#define PHYS_PFN_OFFSET ((unsigned long)(PHYS_OFFSET >> PAGE_SHIFT))
static inline phys_addr_t __virt_to_phys_nodebug(unsigned long x)
{
return (phys_addr_t)x - PAGE_OFFSET + PHYS_OFFSET;
}
static inline unsigned long __phys_to_virt(phys_addr_t x)
{
return x - PHYS_OFFSET + PAGE_OFFSET;
}
#endif
ARM: Better virt_to_page() handling virt_to_page() is incredibly inefficient when virt-to-phys patching is enabled. This is because we end up with this calculation: page = &mem_map[asm virt_to_phys(addr) >> 12 - __pv_phys_offset >> 12] in assembly. The asm virt_to_phys() is equivalent this this operation: addr - PAGE_OFFSET + __pv_phys_offset and we can see that because this is assembly, the compiler has no chance to optimise some of that away. This should reduce down to: page = &mem_map[(addr - PAGE_OFFSET) >> 12] for the common cases. Permit the compiler to make this optimisation by giving it more of the information it needs - do this by providing a virt_to_pfn() macro. Another issue which makes this more complex is that __pv_phys_offset is a 64-bit type on all platforms. This is needlessly wasteful - if we store the physical offset as a PFN, we can save a lot of work having to deal with 64-bit values, which sometimes ends up producing incredibly horrid code: a4c: e3009000 movw r9, #0 a4c: R_ARM_MOVW_ABS_NC __pv_phys_offset a50: e3409000 movt r9, #0 ; r9 = &__pv_phys_offset a50: R_ARM_MOVT_ABS __pv_phys_offset a54: e3002000 movw r2, #0 a54: R_ARM_MOVW_ABS_NC __pv_phys_offset a58: e3402000 movt r2, #0 ; r2 = &__pv_phys_offset a58: R_ARM_MOVT_ABS __pv_phys_offset a5c: e5999004 ldr r9, [r9, #4] ; r9 = high word of __pv_phys_offset a60: e3001000 movw r1, #0 a60: R_ARM_MOVW_ABS_NC mem_map a64: e592c000 ldr ip, [r2] ; ip = low word of __pv_phys_offset Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-03-26 04:45:31 +09:00
#define virt_to_pfn(kaddr) \
((((unsigned long)(kaddr) - PAGE_OFFSET) >> PAGE_SHIFT) + \
PHYS_PFN_OFFSET)
#define __pa_symbol_nodebug(x) __virt_to_phys_nodebug((x))
#ifdef CONFIG_DEBUG_VIRTUAL
extern phys_addr_t __virt_to_phys(unsigned long x);
extern phys_addr_t __phys_addr_symbol(unsigned long x);
#else
#define __virt_to_phys(x) __virt_to_phys_nodebug(x)
#define __phys_addr_symbol(x) __pa_symbol_nodebug(x)
#endif
/*
* These are *only* valid on the kernel direct mapped RAM memory.
* Note: Drivers should NOT use these. They are the wrong
* translation for translating DMA addresses. Use the driver
* DMA support - see dma-mapping.h.
*/
#define virt_to_phys virt_to_phys
static inline phys_addr_t virt_to_phys(const volatile void *x)
{
return __virt_to_phys((unsigned long)(x));
}
#define phys_to_virt phys_to_virt
static inline void *phys_to_virt(phys_addr_t x)
{
return (void *)__phys_to_virt(x);
}
/*
* Drivers should NOT use these either.
*/
#define __pa(x) __virt_to_phys((unsigned long)(x))
#define __pa_symbol(x) __phys_addr_symbol(RELOC_HIDE((unsigned long)(x), 0))
#define __va(x) ((void *)__phys_to_virt((phys_addr_t)(x)))
#define pfn_to_kaddr(pfn) __va((phys_addr_t)(pfn) << PAGE_SHIFT)
extern long long arch_phys_to_idmap_offset;
/*
* These are for systems that have a hardware interconnect supported alias
* of physical memory for idmap purposes. Most cases should leave these
* untouched. Note: this can only return addresses less than 4GiB.
*/
static inline bool arm_has_idmap_alias(void)
{
return IS_ENABLED(CONFIG_MMU) && arch_phys_to_idmap_offset != 0;
}
#define IDMAP_INVALID_ADDR ((u32)~0)
static inline unsigned long phys_to_idmap(phys_addr_t addr)
{
if (IS_ENABLED(CONFIG_MMU) && arch_phys_to_idmap_offset) {
addr += arch_phys_to_idmap_offset;
if (addr > (u32)~0)
addr = IDMAP_INVALID_ADDR;
}
return addr;
}
static inline phys_addr_t idmap_to_phys(unsigned long idmap)
{
phys_addr_t addr = idmap;
if (IS_ENABLED(CONFIG_MMU) && arch_phys_to_idmap_offset)
addr -= arch_phys_to_idmap_offset;
return addr;
}
static inline unsigned long __virt_to_idmap(unsigned long x)
{
return phys_to_idmap(__virt_to_phys(x));
}
#define virt_to_idmap(x) __virt_to_idmap((unsigned long)(x))
/*
* Virtual <-> DMA view memory address translations
* Again, these are *only* valid on the kernel direct mapped RAM
* memory. Use of these is *deprecated* (and that doesn't mean
* use the __ prefixed forms instead.) See dma-mapping.h.
*/
#ifndef __virt_to_bus
#define __virt_to_bus __virt_to_phys
#define __bus_to_virt __phys_to_virt
#define __pfn_to_bus(x) __pfn_to_phys(x)
#define __bus_to_pfn(x) __phys_to_pfn(x)
#endif
/*
* Conversion between a struct page and a physical address.
*
* page_to_pfn(page) convert a struct page * to a PFN number
* pfn_to_page(pfn) convert a _valid_ PFN number to struct page *
*
* virt_to_page(k) convert a _valid_ virtual address to struct page *
* virt_addr_valid(k) indicates whether a virtual address is valid
*/
#define ARCH_PFN_OFFSET PHYS_PFN_OFFSET
ARM: Better virt_to_page() handling virt_to_page() is incredibly inefficient when virt-to-phys patching is enabled. This is because we end up with this calculation: page = &mem_map[asm virt_to_phys(addr) >> 12 - __pv_phys_offset >> 12] in assembly. The asm virt_to_phys() is equivalent this this operation: addr - PAGE_OFFSET + __pv_phys_offset and we can see that because this is assembly, the compiler has no chance to optimise some of that away. This should reduce down to: page = &mem_map[(addr - PAGE_OFFSET) >> 12] for the common cases. Permit the compiler to make this optimisation by giving it more of the information it needs - do this by providing a virt_to_pfn() macro. Another issue which makes this more complex is that __pv_phys_offset is a 64-bit type on all platforms. This is needlessly wasteful - if we store the physical offset as a PFN, we can save a lot of work having to deal with 64-bit values, which sometimes ends up producing incredibly horrid code: a4c: e3009000 movw r9, #0 a4c: R_ARM_MOVW_ABS_NC __pv_phys_offset a50: e3409000 movt r9, #0 ; r9 = &__pv_phys_offset a50: R_ARM_MOVT_ABS __pv_phys_offset a54: e3002000 movw r2, #0 a54: R_ARM_MOVW_ABS_NC __pv_phys_offset a58: e3402000 movt r2, #0 ; r2 = &__pv_phys_offset a58: R_ARM_MOVT_ABS __pv_phys_offset a5c: e5999004 ldr r9, [r9, #4] ; r9 = high word of __pv_phys_offset a60: e3001000 movw r1, #0 a60: R_ARM_MOVW_ABS_NC mem_map a64: e592c000 ldr ip, [r2] ; ip = low word of __pv_phys_offset Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-03-26 04:45:31 +09:00
#define virt_to_page(kaddr) pfn_to_page(virt_to_pfn(kaddr))
#define virt_addr_valid(kaddr) (((unsigned long)(kaddr) >= PAGE_OFFSET && (unsigned long)(kaddr) < (unsigned long)high_memory) \
ARM: Better virt_to_page() handling virt_to_page() is incredibly inefficient when virt-to-phys patching is enabled. This is because we end up with this calculation: page = &mem_map[asm virt_to_phys(addr) >> 12 - __pv_phys_offset >> 12] in assembly. The asm virt_to_phys() is equivalent this this operation: addr - PAGE_OFFSET + __pv_phys_offset and we can see that because this is assembly, the compiler has no chance to optimise some of that away. This should reduce down to: page = &mem_map[(addr - PAGE_OFFSET) >> 12] for the common cases. Permit the compiler to make this optimisation by giving it more of the information it needs - do this by providing a virt_to_pfn() macro. Another issue which makes this more complex is that __pv_phys_offset is a 64-bit type on all platforms. This is needlessly wasteful - if we store the physical offset as a PFN, we can save a lot of work having to deal with 64-bit values, which sometimes ends up producing incredibly horrid code: a4c: e3009000 movw r9, #0 a4c: R_ARM_MOVW_ABS_NC __pv_phys_offset a50: e3409000 movt r9, #0 ; r9 = &__pv_phys_offset a50: R_ARM_MOVT_ABS __pv_phys_offset a54: e3002000 movw r2, #0 a54: R_ARM_MOVW_ABS_NC __pv_phys_offset a58: e3402000 movt r2, #0 ; r2 = &__pv_phys_offset a58: R_ARM_MOVT_ABS __pv_phys_offset a5c: e5999004 ldr r9, [r9, #4] ; r9 = high word of __pv_phys_offset a60: e3001000 movw r1, #0 a60: R_ARM_MOVW_ABS_NC mem_map a64: e592c000 ldr ip, [r2] ; ip = low word of __pv_phys_offset Reviewed-by: Nicolas Pitre <nico@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-03-26 04:45:31 +09:00
&& pfn_valid(virt_to_pfn(kaddr)))
#endif
#include <asm-generic/memory_model.h>
#endif