AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-12-28 11:18:45 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
a2c763e074
linux_dsm_epyc7002/arch/mips/include/asm/pgtable-bits.h

263 lines
8.8 KiB
C
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
* Copyright (C) 1994 - 2002 by Ralf Baechle
* Copyright (C) 1999, 2000, 2001 Silicon Graphics, Inc.
* Copyright (C) 2002 Maciej W. Rozycki
*/
#ifndef _ASM_PGTABLE_BITS_H
#define _ASM_PGTABLE_BITS_H
/*
* Note that we shift the lower 32bits of each EntryLo[01] entry
* 6 bits to the left. That way we can convert the PFN into the
* physical address by a single 'and' operation and gain 6 additional
* bits for storing information which isn't present in a normal
* MIPS page table.
*
* Similar to the Alpha port, we need to keep track of the ref
* and mod bits in software. We have a software "yeah you can read
* from this page" bit, and a hardware one which actually lets the
* process read from the page. On the same token we have a software
* writable bit and the real hardware one which actually lets the
* process write to the page, this keeps a mod bit via the hardware
* dirty bit.
*
* Certain revisions of the R4000 and R5000 have a bug where if a
* certain sequence occurs in the last 3 instructions of an executable
* page, and the following page is not mapped, the cpu can do
* unpredictable things. The code (when it is written) to deal with
* this problem will be in the update_mmu_cache() code for the r4k.
*/
[MIPS] All MIPS32 processors support64-bit physical addresses. Still, only the 4K may actually implement it. Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-09-19 06:46:32 +07:00
#if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
MIPS: tlbex: Better debug output. Pgtable bits are assigned dynamically depending on processor feature and statically based on kernel configuration. To make sense out of the disassembled TLB exception handlers a list of the actual assignments used for a particular configuration and hardware setup can be very useful. Output the actual TLB exception handlers in a format that simplifies their post processsing from dmesg output. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2012-10-17 03:20:26 +07:00
/*
* The following bits are directly used by the TLB hardware
*/
#define _PAGE_R4KBUG (1 << 0) /* workaround for r4k bug */
#define _PAGE_GLOBAL (1 << 0)
#define _PAGE_VALID_SHIFT 1
#define _PAGE_VALID (1 << _PAGE_VALID_SHIFT)
#define _PAGE_SILENT_READ (1 << 1) /* synonym */
#define _PAGE_DIRTY_SHIFT 2
#define _PAGE_DIRTY (1 << _PAGE_DIRTY_SHIFT) /* The MIPS dirty bit */
#define _PAGE_SILENT_WRITE (1 << 2)
#define _CACHE_SHIFT 3
#define _CACHE_MASK (7 << 3)
/*
* The following bits are implemented in software
*
* _PAGE_FILE semantics: set:pagecache unset:swap
*/
#define _PAGE_PRESENT_SHIFT 6
#define _PAGE_PRESENT (1 << _PAGE_PRESENT_SHIFT)
#define _PAGE_READ_SHIFT 7
#define _PAGE_READ (1 << _PAGE_READ_SHIFT)
#define _PAGE_WRITE_SHIFT 8
#define _PAGE_WRITE (1 << _PAGE_WRITE_SHIFT)
#define _PAGE_ACCESSED_SHIFT 9
#define _PAGE_ACCESSED (1 << _PAGE_ACCESSED_SHIFT)
#define _PAGE_MODIFIED_SHIFT 10
#define _PAGE_MODIFIED (1 << _PAGE_MODIFIED_SHIFT)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
MIPS: tlbex: Better debug output. Pgtable bits are assigned dynamically depending on processor feature and statically based on kernel configuration. To make sense out of the disassembled TLB exception handlers a list of the actual assignments used for a particular configuration and hardware setup can be very useful. Output the actual TLB exception handlers in a format that simplifies their post processsing from dmesg output. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2012-10-17 03:20:26 +07:00
#define _PAGE_FILE (1 << 10)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
#elif defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
MIPS: tlbex: Better debug output. Pgtable bits are assigned dynamically depending on processor feature and statically based on kernel configuration. To make sense out of the disassembled TLB exception handlers a list of the actual assignments used for a particular configuration and hardware setup can be very useful. Output the actual TLB exception handlers in a format that simplifies their post processsing from dmesg output. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2012-10-17 03:20:26 +07:00
/*
* The following are implemented by software
*
* _PAGE_FILE semantics: set:pagecache unset:swap
*/
#define _PAGE_PRESENT_SHIFT 0
#define _PAGE_PRESENT (1 << _PAGE_PRESENT_SHIFT)
#define _PAGE_READ_SHIFT 1
#define _PAGE_READ (1 << _PAGE_READ_SHIFT)
#define _PAGE_WRITE_SHIFT 2
#define _PAGE_WRITE (1 << _PAGE_WRITE_SHIFT)
#define _PAGE_ACCESSED_SHIFT 3
#define _PAGE_ACCESSED (1 << _PAGE_ACCESSED_SHIFT)
#define _PAGE_MODIFIED_SHIFT 4
#define _PAGE_MODIFIED (1 << _PAGE_MODIFIED_SHIFT)
#define _PAGE_FILE_SHIFT 4
#define _PAGE_FILE (1 << _PAGE_FILE_SHIFT)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
MIPS: tlbex: Better debug output. Pgtable bits are assigned dynamically depending on processor feature and statically based on kernel configuration. To make sense out of the disassembled TLB exception handlers a list of the actual assignments used for a particular configuration and hardware setup can be very useful. Output the actual TLB exception handlers in a format that simplifies their post processsing from dmesg output. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2012-10-17 03:20:26 +07:00
/*
* And these are the hardware TLB bits
*/
#define _PAGE_GLOBAL_SHIFT 8
#define _PAGE_GLOBAL (1 << _PAGE_GLOBAL_SHIFT)
#define _PAGE_VALID_SHIFT 9
#define _PAGE_VALID (1 << _PAGE_VALID_SHIFT)
#define _PAGE_SILENT_READ (1 << _PAGE_VALID_SHIFT) /* synonym */
#define _PAGE_DIRTY_SHIFT 10
#define _PAGE_DIRTY (1 << _PAGE_DIRTY_SHIFT)
#define _PAGE_SILENT_WRITE (1 << _PAGE_DIRTY_SHIFT)
#define _CACHE_UNCACHED_SHIFT 11
#define _CACHE_UNCACHED (1 << _CACHE_UNCACHED_SHIFT)
#define _CACHE_MASK (1 << _CACHE_UNCACHED_SHIFT)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
#else /* 'Normal' r4K case */
/*
* When using the RI/XI bit support, we have 13 bits of flags below
* the physical address. The RI/XI bits are placed such that a SRL 5
* can strip off the software bits, then a ROTR 2 can move the RI/XI
* into bits [63:62]. This also limits physical address to 56 bits,
* which is more than we need right now.
*/
MIPS: tlbex: Better debug output. Pgtable bits are assigned dynamically depending on processor feature and statically based on kernel configuration. To make sense out of the disassembled TLB exception handlers a list of the actual assignments used for a particular configuration and hardware setup can be very useful. Output the actual TLB exception handlers in a format that simplifies their post processsing from dmesg output. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2012-10-17 03:20:26 +07:00
/*
* The following bits are implemented in software
*
* _PAGE_READ / _PAGE_READ_SHIFT should be unused if cpu_has_rixi.
* _PAGE_FILE semantics: set:pagecache unset:swap
*/
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
#define _PAGE_PRESENT_SHIFT (0)
#define _PAGE_PRESENT (1 << _PAGE_PRESENT_SHIFT)
MIPS: Replace 'kernel_uses_smartmips_rixi' with 'cpu_has_rixi'. Remove usage of the 'kernel_uses_smartmips_rixi' macro from all files and use new 'cpu_has_rixi' instead. Signed-off-by: Steven J. Hill <sjhill@mips.com> Acked-by: David Daney <david.daney@cavium.com>
2012-09-14 04:51:46 +07:00
#define _PAGE_READ_SHIFT (cpu_has_rixi ? _PAGE_PRESENT_SHIFT : _PAGE_PRESENT_SHIFT + 1)
#define _PAGE_READ ({BUG_ON(cpu_has_rixi); 1 << _PAGE_READ_SHIFT; })
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
#define _PAGE_WRITE_SHIFT (_PAGE_READ_SHIFT + 1)
#define _PAGE_WRITE (1 << _PAGE_WRITE_SHIFT)
#define _PAGE_ACCESSED_SHIFT (_PAGE_WRITE_SHIFT + 1)
#define _PAGE_ACCESSED (1 << _PAGE_ACCESSED_SHIFT)
#define _PAGE_MODIFIED_SHIFT (_PAGE_ACCESSED_SHIFT + 1)
#define _PAGE_MODIFIED (1 << _PAGE_MODIFIED_SHIFT)
#define _PAGE_FILE (_PAGE_MODIFIED)
#ifdef CONFIG_HUGETLB_PAGE
/* huge tlb page */
#define _PAGE_HUGE_SHIFT (_PAGE_MODIFIED_SHIFT + 1)
#define _PAGE_HUGE (1 << _PAGE_HUGE_SHIFT)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
#else
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
#define _PAGE_HUGE_SHIFT (_PAGE_MODIFIED_SHIFT)
#define _PAGE_HUGE ({BUG(); 1; }) /* Dummy value */
#endif
[MIPS] Tidy up cache attributes Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-09-19 06:51:57 +07:00
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
/* Page cannot be executed */
MIPS: Replace 'kernel_uses_smartmips_rixi' with 'cpu_has_rixi'. Remove usage of the 'kernel_uses_smartmips_rixi' macro from all files and use new 'cpu_has_rixi' instead. Signed-off-by: Steven J. Hill <sjhill@mips.com> Acked-by: David Daney <david.daney@cavium.com>
2012-09-14 04:51:46 +07:00
#define _PAGE_NO_EXEC_SHIFT (cpu_has_rixi ? _PAGE_HUGE_SHIFT + 1 : _PAGE_HUGE_SHIFT)
#define _PAGE_NO_EXEC ({BUG_ON(!cpu_has_rixi); 1 << _PAGE_NO_EXEC_SHIFT; })
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
/* Page cannot be read */
MIPS: Replace 'kernel_uses_smartmips_rixi' with 'cpu_has_rixi'. Remove usage of the 'kernel_uses_smartmips_rixi' macro from all files and use new 'cpu_has_rixi' instead. Signed-off-by: Steven J. Hill <sjhill@mips.com> Acked-by: David Daney <david.daney@cavium.com>
2012-09-14 04:51:46 +07:00
#define _PAGE_NO_READ_SHIFT (cpu_has_rixi ? _PAGE_NO_EXEC_SHIFT + 1 : _PAGE_NO_EXEC_SHIFT)
#define _PAGE_NO_READ ({BUG_ON(!cpu_has_rixi); 1 << _PAGE_NO_READ_SHIFT; })
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
#define _PAGE_GLOBAL_SHIFT (_PAGE_NO_READ_SHIFT + 1)
#define _PAGE_GLOBAL (1 << _PAGE_GLOBAL_SHIFT)
#define _PAGE_VALID_SHIFT (_PAGE_GLOBAL_SHIFT + 1)
#define _PAGE_VALID (1 << _PAGE_VALID_SHIFT)
/* synonym */
#define _PAGE_SILENT_READ (_PAGE_VALID)
/* The MIPS dirty bit */
#define _PAGE_DIRTY_SHIFT (_PAGE_VALID_SHIFT + 1)
#define _PAGE_DIRTY (1 << _PAGE_DIRTY_SHIFT)
#define _PAGE_SILENT_WRITE (_PAGE_DIRTY)
#define _CACHE_SHIFT (_PAGE_DIRTY_SHIFT + 1)
#define _CACHE_MASK (7 << _CACHE_SHIFT)
#define _PFN_SHIFT (PAGE_SHIFT - 12 + _CACHE_SHIFT + 3)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
[MIPS] Tidy up cache attributes Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-09-19 06:51:57 +07:00
#endif /* defined(CONFIG_64BIT_PHYS_ADDR && defined(CONFIG_CPU_MIPS32) */
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
#ifndef _PFN_SHIFT
#define _PFN_SHIFT PAGE_SHIFT
#endif
#define _PFN_MASK (~((1 << (_PFN_SHIFT)) - 1))
#ifndef _PAGE_NO_READ
#define _PAGE_NO_READ ({BUG(); 0; })
#define _PAGE_NO_READ_SHIFT ({BUG(); 0; })
#endif
#ifndef _PAGE_NO_EXEC
#define _PAGE_NO_EXEC ({BUG(); 0; })
#endif
#ifndef _PAGE_GLOBAL_SHIFT
#define _PAGE_GLOBAL_SHIFT ilog2(_PAGE_GLOBAL)
#endif
#ifndef __ASSEMBLY__
/*
* pte_to_entrylo converts a page table entry (PTE) into a Mips
* entrylo0/1 value.
*/
static inline uint64_t pte_to_entrylo(unsigned long pte_val)
{
MIPS: Replace 'kernel_uses_smartmips_rixi' with 'cpu_has_rixi'. Remove usage of the 'kernel_uses_smartmips_rixi' macro from all files and use new 'cpu_has_rixi' instead. Signed-off-by: Steven J. Hill <sjhill@mips.com> Acked-by: David Daney <david.daney@cavium.com>
2012-09-14 04:51:46 +07:00
if (cpu_has_rixi) {
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
int sa;
#ifdef CONFIG_32BIT
sa = 31 - _PAGE_NO_READ_SHIFT;
#else
sa = 63 - _PAGE_NO_READ_SHIFT;
#endif
/*
* C has no way to express that this is a DSRL
* _PAGE_NO_EXEC_SHIFT followed by a ROTR 2. Luckily
* in the fast path this is done in assembly
*/
return (pte_val >> _PAGE_GLOBAL_SHIFT) |
((pte_val & (_PAGE_NO_EXEC | _PAGE_NO_READ)) << sa);
}
return pte_val >> _PAGE_GLOBAL_SHIFT;
}
#endif
[MIPS] Tidy up cache attributes Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-09-19 06:51:57 +07:00
/*
* Cache attributes
*/
#if defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)
#define _CACHE_CACHABLE_NONCOHERENT 0
#elif defined(CONFIG_CPU_SB1)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
/* No penalty for being coherent on the SB1, so just
use it for "noncoherent" spaces, too. Shouldn't hurt. */
[MIPS] Tidy up cache attributes Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-09-19 06:51:57 +07:00
#define _CACHE_UNCACHED (2<<_CACHE_SHIFT)
#define _CACHE_CACHABLE_COW (5<<_CACHE_SHIFT)
#define _CACHE_CACHABLE_NONCOHERENT (5<<_CACHE_SHIFT)
#define _CACHE_UNCACHED_ACCELERATED (7<<_CACHE_SHIFT)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
#elif defined(CONFIG_CPU_RM9000)
[MIPS] Tidy up cache attributes Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-09-19 06:51:57 +07:00
#define _CACHE_WT (0<<_CACHE_SHIFT)
#define _CACHE_WTWA (1<<_CACHE_SHIFT)
#define _CACHE_UC_B (2<<_CACHE_SHIFT)
#define _CACHE_WB (3<<_CACHE_SHIFT)
#define _CACHE_CWBEA (4<<_CACHE_SHIFT)
#define _CACHE_CWB (5<<_CACHE_SHIFT)
#define _CACHE_UCNB (6<<_CACHE_SHIFT)
#define _CACHE_FPC (7<<_CACHE_SHIFT)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
[MIPS] Tidy up cache attributes Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-09-19 06:51:57 +07:00
#define _CACHE_UNCACHED _CACHE_UC_B
#define _CACHE_CACHABLE_NONCOHERENT _CACHE_WB
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
#else
[MIPS] Tidy up cache attributes Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2007-09-19 06:51:57 +07:00
#define _CACHE_CACHABLE_NO_WA (0<<_CACHE_SHIFT) /* R4600 only */
#define _CACHE_CACHABLE_WA (1<<_CACHE_SHIFT) /* R4600 only */
#define _CACHE_UNCACHED (2<<_CACHE_SHIFT) /* R4[0246]00 */
#define _CACHE_CACHABLE_NONCOHERENT (3<<_CACHE_SHIFT) /* R4[0246]00 */
#define _CACHE_CACHABLE_CE (4<<_CACHE_SHIFT) /* R4[04]00MC only */
#define _CACHE_CACHABLE_COW (5<<_CACHE_SHIFT) /* R4[04]00MC only */
#define _CACHE_CACHABLE_COHERENT (5<<_CACHE_SHIFT) /* MIPS32R2 CMP */
#define _CACHE_CACHABLE_CUW (6<<_CACHE_SHIFT) /* R4[04]00MC only */
#define _CACHE_UNCACHED_ACCELERATED (7<<_CACHE_SHIFT) /* R10000 only */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
#endif
MIPS: Replace 'kernel_uses_smartmips_rixi' with 'cpu_has_rixi'. Remove usage of the 'kernel_uses_smartmips_rixi' macro from all files and use new 'cpu_has_rixi' instead. Signed-off-by: Steven J. Hill <sjhill@mips.com> Acked-by: David Daney <david.daney@cavium.com>
2012-09-14 04:51:46 +07:00
#define __READABLE (_PAGE_SILENT_READ | _PAGE_ACCESSED | (cpu_has_rixi ? 0 : _PAGE_READ))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
#define __WRITEABLE (_PAGE_WRITE | _PAGE_SILENT_WRITE | _PAGE_MODIFIED)
MIPS: Implement Read Inhibit/eXecute Inhibit The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit (XI) bits in the page tables work. The upper two bits of EntryLo{0,1} are RI and XI when the feature is enabled in the PageGrain register. SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to 64-bit systems by continuing to place the RI and XI bits in the top of EntryLo even when EntryLo is 64-bits wide. Because we need to carry the RI and XI bits in the PTE, the layout of the PTE is changed. There is a two instruction overhead in the TLB refill hot path to get the EntryLo bits into the proper position. Also the TLB load exception has to probe the TLB to check if RI or XI caused the exception. Also of note is that the layout of the PTE bits is done at compile and runtime rather than statically. In the 32-bit case this allows for the same number of PFN bits as before the patch as the _PAGE_HUGE is not supported in 32-bit kernels (we have _PAGE_NO_EXEC and _PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE). The patch is tested on Cavium Octeon+, but should also work on 32-bit systems with the Smart-MIPS ASE. Signed-off-by: David Daney <ddaney@caviumnetworks.com> To: linux-mips@linux-mips.org Patchwork: http://patchwork.linux-mips.org/patch/952/ Patchwork: http://patchwork.linux-mips.org/patch/956/ Patchwork: http://patchwork.linux-mips.org/patch/962/ Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2010-02-11 06:12:47 +07:00
#define _PAGE_CHG_MASK (_PFN_MASK | _PAGE_ACCESSED | _PAGE_MODIFIED | _CACHE_MASK)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 05:20:36 +07:00
#endif /* _ASM_PGTABLE_BITS_H */
Reference in New Issue Copy Permalink