Resguardo inicial

linux-7.2.3/arch/powerpc/mm/nohash/44x.c

@@ -0,0 +1,242 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Modifications by Matt Porter (mporter@mvista.com) to support
+ * PPC44x Book E processors.
+ *
+ * This file contains the routines for initializing the MMU
+ * on the 4xx series of chips.
+ *  -- paulus
+ *
+ *  Derived from arch/ppc/mm/init.c:
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au)
+ *  and Cort Dougan (PReP) (cort@cs.nmt.edu)
+ *    Copyright (C) 1996 Paul Mackerras
+ *
+ *  Derived from "arch/i386/mm/init.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ */
+
+#include <linux/init.h>
+#include <linux/memblock.h>
+
+#include <asm/mmu.h>
+#include <asm/page.h>
+#include <asm/cacheflush.h>
+#include <asm/text-patching.h>
+#include <asm/smp.h>
+
+#include <mm/mmu_decl.h>
+
+/* Used by the 44x TLB replacement exception handler.
+ * Just needed it declared someplace.
+ */
+unsigned int tlb_44x_index; /* = 0 */
+unsigned int tlb_44x_hwater = PPC44x_TLB_SIZE - 1 - PPC44x_EARLY_TLBS;
+int icache_44x_need_flush;
+
+unsigned long tlb_47x_boltmap[1024/8];
+
+static void __init ppc44x_update_tlb_hwater(void)
+{
+	/* The TLB miss handlers hard codes the watermark in a cmpli
+	 * instruction to improve performances rather than loading it
+	 * from the global variable. Thus, we patch the instructions
+	 * in the 2 TLB miss handlers when updating the value
+	 */
+	modify_instruction_site(&patch__tlb_44x_hwater_D, 0xffff, tlb_44x_hwater);
+	modify_instruction_site(&patch__tlb_44x_hwater_I, 0xffff, tlb_44x_hwater);
+}
+
+/*
+ * "Pins" a 256MB TLB entry in AS0 for kernel lowmem for 44x type MMU
+ */
+static void __init ppc44x_pin_tlb(unsigned int virt, unsigned int phys)
+{
+	unsigned int entry = tlb_44x_hwater--;
+
+	ppc44x_update_tlb_hwater();
+
+	mtspr(SPRN_MMUCR, 0);
+
+	__asm__ __volatile__(
+		"tlbwe	%2,%3,%4\n"
+		"tlbwe	%1,%3,%5\n"
+		"tlbwe	%0,%3,%6\n"
+	:
+	: "r" (PPC44x_TLB_SW | PPC44x_TLB_SR | PPC44x_TLB_SX | PPC44x_TLB_G),
+	  "r" (phys),
+	  "r" (virt | PPC44x_TLB_VALID | PPC44x_TLB_256M),
+	  "r" (entry),
+	  "i" (PPC44x_TLB_PAGEID),
+	  "i" (PPC44x_TLB_XLAT),
+	  "i" (PPC44x_TLB_ATTRIB));
+}
+
+static int __init ppc47x_find_free_bolted(void)
+{
+	unsigned int mmube0 = mfspr(SPRN_MMUBE0);
+	unsigned int mmube1 = mfspr(SPRN_MMUBE1);
+
+	if (!(mmube0 & MMUBE0_VBE0))
+		return 0;
+	if (!(mmube0 & MMUBE0_VBE1))
+		return 1;
+	if (!(mmube0 & MMUBE0_VBE2))
+		return 2;
+	if (!(mmube1 & MMUBE1_VBE3))
+		return 3;
+	if (!(mmube1 & MMUBE1_VBE4))
+		return 4;
+	if (!(mmube1 & MMUBE1_VBE5))
+		return 5;
+	return -1;
+}
+
+static void __init ppc47x_update_boltmap(void)
+{
+	unsigned int mmube0 = mfspr(SPRN_MMUBE0);
+	unsigned int mmube1 = mfspr(SPRN_MMUBE1);
+
+	if (mmube0 & MMUBE0_VBE0)
+		__set_bit((mmube0 >> MMUBE0_IBE0_SHIFT) & 0xff,
+			  tlb_47x_boltmap);
+	if (mmube0 & MMUBE0_VBE1)
+		__set_bit((mmube0 >> MMUBE0_IBE1_SHIFT) & 0xff,
+			  tlb_47x_boltmap);
+	if (mmube0 & MMUBE0_VBE2)
+		__set_bit((mmube0 >> MMUBE0_IBE2_SHIFT) & 0xff,
+			  tlb_47x_boltmap);
+	if (mmube1 & MMUBE1_VBE3)
+		__set_bit((mmube1 >> MMUBE1_IBE3_SHIFT) & 0xff,
+			  tlb_47x_boltmap);
+	if (mmube1 & MMUBE1_VBE4)
+		__set_bit((mmube1 >> MMUBE1_IBE4_SHIFT) & 0xff,
+			  tlb_47x_boltmap);
+	if (mmube1 & MMUBE1_VBE5)
+		__set_bit((mmube1 >> MMUBE1_IBE5_SHIFT) & 0xff,
+			  tlb_47x_boltmap);
+}
+
+/*
+ * "Pins" a 256MB TLB entry in AS0 for kernel lowmem for 47x type MMU
+ */
+static void __init ppc47x_pin_tlb(unsigned int virt, unsigned int phys)
+{
+	unsigned int rA;
+	int bolted;
+
+	/* Base rA is HW way select, way 0, bolted bit set */
+	rA = 0x88000000;
+
+	/* Look for a bolted entry slot */
+	bolted = ppc47x_find_free_bolted();
+	BUG_ON(bolted < 0);
+
+	/* Insert bolted slot number */
+	rA |= bolted << 24;
+
+	pr_debug("256M TLB entry for 0x%08x->0x%08x in bolt slot %d\n",
+		 virt, phys, bolted);
+
+	mtspr(SPRN_MMUCR, 0);
+
+	__asm__ __volatile__(
+		"tlbwe	%2,%3,0\n"
+		"tlbwe	%1,%3,1\n"
+		"tlbwe	%0,%3,2\n"
+		:
+		: "r" (PPC47x_TLB2_SW | PPC47x_TLB2_SR |
+		       PPC47x_TLB2_SX
+#ifdef CONFIG_SMP
+		       | PPC47x_TLB2_M
+#endif
+		       ),
+		  "r" (phys),
+		  "r" (virt | PPC47x_TLB0_VALID | PPC47x_TLB0_256M),
+		  "r" (rA));
+}
+
+void __init MMU_init_hw(void)
+{
+	/* This is not useful on 47x but won't hurt either */
+	ppc44x_update_tlb_hwater();
+
+	flush_instruction_cache();
+}
+
+unsigned long __init mmu_mapin_ram(unsigned long base, unsigned long top)
+{
+	unsigned long addr;
+	unsigned long memstart = memstart_addr & ~(PPC_PIN_SIZE - 1);
+
+	/* Pin in enough TLBs to cover any lowmem not covered by the
+	 * initial 256M mapping established in head_44x.S */
+	for (addr = memstart + PPC_PIN_SIZE; addr < lowmem_end_addr;
+	     addr += PPC_PIN_SIZE) {
+		if (mmu_has_feature(MMU_FTR_TYPE_47x))
+			ppc47x_pin_tlb(addr + PAGE_OFFSET, addr);
+		else
+			ppc44x_pin_tlb(addr + PAGE_OFFSET, addr);
+	}
+	if (mmu_has_feature(MMU_FTR_TYPE_47x)) {
+		ppc47x_update_boltmap();
+
+#ifdef DEBUG
+		{
+			int i;
+
+			printk(KERN_DEBUG "bolted entries: ");
+			for (i = 0; i < 255; i++) {
+				if (test_bit(i, tlb_47x_boltmap))
+					printk("%d ", i);
+			}
+			printk("\n");
+		}
+#endif /* DEBUG */
+	}
+	return total_lowmem;
+}
+
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	u64 size;
+
+#ifndef CONFIG_NONSTATIC_KERNEL
+	/* We don't currently support the first MEMBLOCK not mapping 0
+	 * physical on those processors
+	 */
+	BUG_ON(first_memblock_base != 0);
+#endif
+
+	/* 44x has a 256M TLB entry pinned at boot */
+	size = (min_t(u64, first_memblock_size, PPC_PIN_SIZE));
+	memblock_set_current_limit(first_memblock_base + size);
+}
+
+#ifdef CONFIG_SMP
+void __init mmu_init_secondary(int cpu)
+{
+	unsigned long addr;
+	unsigned long memstart = memstart_addr & ~(PPC_PIN_SIZE - 1);
+
+	/* Pin in enough TLBs to cover any lowmem not covered by the
+	 * initial 256M mapping established in head_44x.S
+	 *
+	 * WARNING: This is called with only the first 256M of the
+	 * linear mapping in the TLB and we can't take faults yet
+	 * so beware of what this code uses. It runs off a temporary
+	 * stack. current (r2) isn't initialized, smp_processor_id()
+	 * will not work, current thread info isn't accessible, ...
+	 */
+	for (addr = memstart + PPC_PIN_SIZE; addr < lowmem_end_addr;
+	     addr += PPC_PIN_SIZE) {
+		if (mmu_has_feature(MMU_FTR_TYPE_47x))
+			ppc47x_pin_tlb(addr + PAGE_OFFSET, addr);
+		else
+			ppc44x_pin_tlb(addr + PAGE_OFFSET, addr);
+	}
+}
+#endif /* CONFIG_SMP */

linux-7.2.3/arch/powerpc/mm/nohash/8xx.c

@@ -0,0 +1,222 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * This file contains the routines for initializing the MMU
+ * on the 8xx series of chips.
+ *  -- christophe
+ *
+ *  Derived from arch/powerpc/mm/40x_mmu.c:
+ */
+
+#include <linux/memblock.h>
+#include <linux/hugetlb.h>
+
+#include <asm/fixmap.h>
+#include <asm/pgalloc.h>
+
+#include <mm/mmu_decl.h>
+
+#define IMMR_SIZE (FIX_IMMR_SIZE << PAGE_SHIFT)
+
+static unsigned long block_mapped_ram;
+
+/*
+ * Return PA for this VA if it is in an area mapped with LTLBs or fixmap.
+ * Otherwise, returns 0
+ */
+phys_addr_t v_block_mapped(unsigned long va)
+{
+	unsigned long p = PHYS_IMMR_BASE;
+
+	if (va >= VIRT_IMMR_BASE && va < VIRT_IMMR_BASE + IMMR_SIZE)
+		return p + va - VIRT_IMMR_BASE;
+	if (va >= PAGE_OFFSET && va < PAGE_OFFSET + block_mapped_ram)
+		return __pa(va);
+	return 0;
+}
+
+/*
+ * Return VA for a given PA mapped with LTLBs or fixmap
+ * Return 0 if not mapped
+ */
+unsigned long p_block_mapped(phys_addr_t pa)
+{
+	unsigned long p = PHYS_IMMR_BASE;
+
+	if (pa >= p && pa < p + IMMR_SIZE)
+		return VIRT_IMMR_BASE + pa - p;
+	if (pa < block_mapped_ram)
+		return (unsigned long)__va(pa);
+	return 0;
+}
+
+static int __ref __early_map_kernel_hugepage(unsigned long va, phys_addr_t pa,
+					     pgprot_t prot, int psize, bool new)
+{
+	pmd_t *pmdp = pmd_off_k(va);
+	pte_t *ptep;
+	unsigned int shift = mmu_psize_to_shift(psize);
+
+	if (new) {
+		if (WARN_ON(slab_is_available()))
+			return -EINVAL;
+
+		if (psize == MMU_PAGE_8M) {
+			if (WARN_ON(!pmd_none(*pmdp) || !pmd_none(*(pmdp + 1))))
+				return -EINVAL;
+
+			ptep = early_alloc_pgtable(PTE_FRAG_SIZE);
+			pmd_populate_kernel(&init_mm, pmdp, ptep);
+
+			ptep = early_alloc_pgtable(PTE_FRAG_SIZE);
+			pmd_populate_kernel(&init_mm, pmdp + 1, ptep);
+
+			ptep = (pte_t *)pmdp;
+		} else {
+			ptep = early_pte_alloc_kernel(pmdp, va);
+			/* The PTE should never be already present */
+			if (WARN_ON(pte_present(*ptep) && pgprot_val(prot)))
+				return -EINVAL;
+		}
+	} else {
+		if (psize == MMU_PAGE_8M)
+			ptep = (pte_t *)pmdp;
+		else
+			ptep = pte_offset_kernel(pmdp, va);
+	}
+
+	if (WARN_ON(!ptep))
+		return -ENOMEM;
+
+	set_huge_pte_at(&init_mm, va, ptep,
+			arch_make_huge_pte(pfn_pte(pa >> PAGE_SHIFT, prot), shift, 0),
+			1UL << shift);
+
+	return 0;
+}
+
+/*
+ * MMU_init_hw does the chip-specific initialization of the MMU hardware.
+ */
+void __init MMU_init_hw(void)
+{
+}
+
+static bool immr_is_mapped __initdata;
+
+void __init mmu_mapin_immr(void)
+{
+	if (immr_is_mapped)
+		return;
+
+	immr_is_mapped = true;
+
+	__early_map_kernel_hugepage(VIRT_IMMR_BASE, PHYS_IMMR_BASE,
+				    PAGE_KERNEL_NCG, MMU_PAGE_512K, true);
+}
+
+static int mmu_mapin_ram_chunk(unsigned long offset, unsigned long top,
+			       pgprot_t prot, bool new)
+{
+	unsigned long v = PAGE_OFFSET + offset;
+	unsigned long p = offset;
+	int err = 0;
+
+	WARN_ON(!IS_ALIGNED(offset, SZ_16K) || !IS_ALIGNED(top, SZ_16K));
+
+	for (; p < ALIGN(p, SZ_512K) && p < top && !err; p += SZ_16K, v += SZ_16K)
+		err = __early_map_kernel_hugepage(v, p, prot, MMU_PAGE_16K, new);
+	for (; p < ALIGN(p, SZ_8M) && p < top && !err; p += SZ_512K, v += SZ_512K)
+		err = __early_map_kernel_hugepage(v, p, prot, MMU_PAGE_512K, new);
+	for (; p < ALIGN_DOWN(top, SZ_8M) && p < top && !err; p += SZ_8M, v += SZ_8M)
+		err = __early_map_kernel_hugepage(v, p, prot, MMU_PAGE_8M, new);
+	for (; p < ALIGN_DOWN(top, SZ_512K) && p < top && !err; p += SZ_512K, v += SZ_512K)
+		err = __early_map_kernel_hugepage(v, p, prot, MMU_PAGE_512K, new);
+	for (; p < ALIGN_DOWN(top, SZ_16K) && p < top && !err; p += SZ_16K, v += SZ_16K)
+		err = __early_map_kernel_hugepage(v, p, prot, MMU_PAGE_16K, new);
+
+	if (!new)
+		flush_tlb_kernel_range(PAGE_OFFSET + v, PAGE_OFFSET + top);
+
+	return err;
+}
+
+unsigned long __init mmu_mapin_ram(unsigned long base, unsigned long top)
+{
+	unsigned long etext8 = ALIGN(__pa(_etext), SZ_8M);
+	unsigned long sinittext = __pa(_sinittext);
+	bool strict_boundary = strict_kernel_rwx_enabled() || debug_pagealloc_enabled_or_kfence();
+	unsigned long boundary = strict_boundary ? sinittext : etext8;
+	unsigned long einittext8 = ALIGN(__pa(_einittext), SZ_8M);
+
+	WARN_ON(top < einittext8);
+
+	mmu_mapin_immr();
+
+	mmu_mapin_ram_chunk(0, boundary, PAGE_KERNEL_X, true);
+	if (debug_pagealloc_enabled_or_kfence()) {
+		top = boundary;
+	} else {
+		mmu_mapin_ram_chunk(boundary, einittext8, PAGE_KERNEL_X, true);
+		mmu_mapin_ram_chunk(einittext8, top, PAGE_KERNEL, true);
+	}
+
+	if (top > SZ_32M)
+		memblock_set_current_limit(top);
+
+	block_mapped_ram = top;
+
+	return top;
+}
+
+int mmu_mark_initmem_nx(void)
+{
+	unsigned long etext8 = ALIGN(__pa(_etext), SZ_8M);
+	unsigned long sinittext = __pa(_sinittext);
+	unsigned long boundary = strict_kernel_rwx_enabled() ? sinittext : etext8;
+	unsigned long einittext8 = ALIGN(__pa(_einittext), SZ_8M);
+	int err = 0;
+
+	if (!debug_pagealloc_enabled_or_kfence())
+		err = mmu_mapin_ram_chunk(boundary, einittext8, PAGE_KERNEL, false);
+
+	if (IS_ENABLED(CONFIG_PIN_TLB_TEXT))
+		mmu_pin_tlb(block_mapped_ram, false);
+
+	return err;
+}
+
+#ifdef CONFIG_STRICT_KERNEL_RWX
+int mmu_mark_rodata_ro(void)
+{
+	unsigned long sinittext = __pa(_sinittext);
+	int err;
+
+	err = mmu_mapin_ram_chunk(0, sinittext, PAGE_KERNEL_ROX, false);
+	if (IS_ENABLED(CONFIG_PIN_TLB_DATA))
+		mmu_pin_tlb(block_mapped_ram, true);
+
+	return err;
+}
+#endif
+
+void __init setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				       phys_addr_t first_memblock_size)
+{
+	/* We don't currently support the first MEMBLOCK not mapping 0
+	 * physical on those processors
+	 */
+	BUG_ON(first_memblock_base != 0);
+
+	/* 8xx can only access 32MB at the moment */
+	memblock_set_current_limit(min_t(u64, first_memblock_size, SZ_32M));
+}
+
+int pud_clear_huge(pud_t *pud)
+{
+	 return 0;
+}
+
+int pmd_clear_huge(pmd_t *pmd)
+{
+	 return 0;
+}

linux-7.2.3/arch/powerpc/mm/nohash/Makefile

@@ -0,0 +1,16 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-y				+= mmu_context.o tlb.o tlb_low.o kup.o
+obj-$(CONFIG_PPC_BOOK3E_64)  	+= tlb_64e.o tlb_low_64e.o book3e_pgtable.o
+obj-$(CONFIG_44x)		+= 44x.o
+obj-$(CONFIG_PPC_8xx)		+= 8xx.o
+obj-$(CONFIG_PPC_E500)		+= e500.o
+obj-$(CONFIG_RANDOMIZE_BASE)	+= kaslr_booke.o
+ifdef CONFIG_HUGETLB_PAGE
+obj-$(CONFIG_PPC_E500)	+= e500_hugetlbpage.o
+endif
+
+# Disable kcov instrumentation on sensitive code
+# This is necessary for booting with kcov enabled on book3e machines
+KCOV_INSTRUMENT_tlb.o := n
+KCOV_INSTRUMENT_e500.o := n

linux-7.2.3/arch/powerpc/mm/nohash/book3e_pgtable.c

@@ -0,0 +1,132 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright 2005, Paul Mackerras, IBM Corporation.
+ * Copyright 2009, Benjamin Herrenschmidt, IBM Corporation.
+ * Copyright 2015-2016, Aneesh Kumar K.V, IBM Corporation.
+ */
+
+#include <linux/sched.h>
+#include <linux/memblock.h>
+#include <asm/pgalloc.h>
+#include <asm/tlb.h>
+#include <asm/dma.h>
+#include <asm/text-patching.h>
+
+#include <mm/mmu_decl.h>
+
+#ifdef CONFIG_SPARSEMEM_VMEMMAP
+/*
+ * On Book3E CPUs, the vmemmap is currently mapped in the top half of
+ * the vmalloc space using normal page tables, though the size of
+ * pages encoded in the PTEs can be different
+ */
+int __meminit vmemmap_create_mapping(unsigned long start,
+				     unsigned long page_size,
+				     unsigned long phys)
+{
+	/* Create a PTE encoding without page size */
+	unsigned long i, flags = _PAGE_PRESENT | _PAGE_ACCESSED |
+		_PAGE_KERNEL_RW;
+
+	/* PTEs only contain page size encodings up to 32M */
+	BUG_ON(mmu_psize_defs[mmu_vmemmap_psize].shift - 10 > 0xf);
+
+	/* Encode the size in the PTE */
+	flags |= (mmu_psize_defs[mmu_vmemmap_psize].shift - 10) << 8;
+
+	/* For each PTE for that area, map things. Note that we don't
+	 * increment phys because all PTEs are of the large size and
+	 * thus must have the low bits clear
+	 */
+	for (i = 0; i < page_size; i += PAGE_SIZE)
+		BUG_ON(map_kernel_page(start + i, phys, __pgprot(flags)));
+
+	return 0;
+}
+
+#ifdef CONFIG_MEMORY_HOTPLUG
+void vmemmap_remove_mapping(unsigned long start,
+			    unsigned long page_size)
+{
+}
+#endif
+#endif /* CONFIG_SPARSEMEM_VMEMMAP */
+
+static void __init *early_alloc_pgtable(unsigned long size)
+{
+	void *ptr;
+
+	ptr = memblock_alloc_try_nid(size, size, MEMBLOCK_LOW_LIMIT,
+				     __pa(MAX_DMA_ADDRESS), NUMA_NO_NODE);
+
+	if (!ptr)
+		panic("%s: Failed to allocate %lu bytes align=0x%lx max_addr=%lx\n",
+		      __func__, size, size, __pa(MAX_DMA_ADDRESS));
+
+	return ptr;
+}
+
+/*
+ * map_kernel_page currently only called by __ioremap
+ * map_kernel_page adds an entry to the ioremap page table
+ * and adds an entry to the HPT, possibly bolting it
+ */
+int __ref map_kernel_page(unsigned long ea, phys_addr_t pa, pgprot_t prot)
+{
+	pgd_t *pgdp;
+	p4d_t *p4dp;
+	pud_t *pudp;
+	pmd_t *pmdp;
+	pte_t *ptep;
+
+	BUILD_BUG_ON(TASK_SIZE_USER64 > PGTABLE_RANGE);
+	if (slab_is_available()) {
+		pgdp = pgd_offset_k(ea);
+		p4dp = p4d_offset(pgdp, ea);
+		pudp = pud_alloc(&init_mm, p4dp, ea);
+		if (!pudp)
+			return -ENOMEM;
+		pmdp = pmd_alloc(&init_mm, pudp, ea);
+		if (!pmdp)
+			return -ENOMEM;
+		ptep = pte_alloc_kernel(pmdp, ea);
+		if (!ptep)
+			return -ENOMEM;
+	} else {
+		pgdp = pgd_offset_k(ea);
+		p4dp = p4d_offset(pgdp, ea);
+		if (p4d_none(*p4dp)) {
+			pudp = early_alloc_pgtable(PUD_TABLE_SIZE);
+			p4d_populate(&init_mm, p4dp, pudp);
+		}
+		pudp = pud_offset(p4dp, ea);
+		if (pud_none(*pudp)) {
+			pmdp = early_alloc_pgtable(PMD_TABLE_SIZE);
+			pud_populate(&init_mm, pudp, pmdp);
+		}
+		pmdp = pmd_offset(pudp, ea);
+		if (!pmd_present(*pmdp)) {
+			ptep = early_alloc_pgtable(PTE_TABLE_SIZE);
+			pmd_populate_kernel(&init_mm, pmdp, ptep);
+		}
+		ptep = pte_offset_kernel(pmdp, ea);
+	}
+	set_pte_at(&init_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT, prot));
+
+	smp_wmb();
+	return 0;
+}
+
+void __patch_exception(int exc, unsigned long addr)
+{
+	unsigned int *ibase = &interrupt_base_book3e;
+
+	/*
+	 * Our exceptions vectors start with a NOP and -then- a branch
+	 * to deal with single stepping from userspace which stops on
+	 * the second instruction. Thus we need to patch the second
+	 * instruction of the exception, not the first one.
+	 */
+
+	patch_branch(ibase + (exc / 4) + 1, addr, 0);
+}

linux-7.2.3/arch/powerpc/mm/nohash/e500.c

@@ -0,0 +1,379 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Modifications by Kumar Gala (galak@kernel.crashing.org) to support
+ * E500 Book E processors.
+ *
+ * Copyright 2004,2010 Freescale Semiconductor, Inc.
+ *
+ * This file contains the routines for initializing the MMU
+ * on the 4xx series of chips.
+ *  -- paulus
+ *
+ *  Derived from arch/ppc/mm/init.c:
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au)
+ *  and Cort Dougan (PReP) (cort@cs.nmt.edu)
+ *    Copyright (C) 1996 Paul Mackerras
+ *
+ *  Derived from "arch/i386/mm/init.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ */
+
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/stddef.h>
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/highmem.h>
+#include <linux/memblock.h>
+#include <linux/of_fdt.h>
+
+#include <asm/io.h>
+#include <asm/mmu_context.h>
+#include <asm/mmu.h>
+#include <linux/uaccess.h>
+#include <asm/smp.h>
+#include <asm/machdep.h>
+#include <asm/setup.h>
+#include <asm/paca.h>
+
+#include <mm/mmu_decl.h>
+
+unsigned int tlbcam_index;
+
+struct tlbcam TLBCAM[NUM_TLBCAMS];
+
+static struct {
+	unsigned long start;
+	unsigned long limit;
+	phys_addr_t phys;
+} tlbcam_addrs[NUM_TLBCAMS];
+
+#ifdef CONFIG_PPC_85xx
+/*
+ * Return PA for this VA if it is mapped by a CAM, or 0
+ */
+phys_addr_t v_block_mapped(unsigned long va)
+{
+	int b;
+	for (b = 0; b < tlbcam_index; ++b)
+		if (va >= tlbcam_addrs[b].start && va < tlbcam_addrs[b].limit)
+			return tlbcam_addrs[b].phys + (va - tlbcam_addrs[b].start);
+	return 0;
+}
+
+/*
+ * Return VA for a given PA or 0 if not mapped
+ */
+unsigned long p_block_mapped(phys_addr_t pa)
+{
+	int b;
+	for (b = 0; b < tlbcam_index; ++b)
+		if (pa >= tlbcam_addrs[b].phys
+			&& pa < (tlbcam_addrs[b].limit-tlbcam_addrs[b].start)
+		              +tlbcam_addrs[b].phys)
+			return tlbcam_addrs[b].start+(pa-tlbcam_addrs[b].phys);
+	return 0;
+}
+#endif
+
+/*
+ * Set up a variable-size TLB entry (tlbcam). The parameters are not checked;
+ * in particular size must be a power of 4 between 4k and the max supported by
+ * an implementation; max may further be limited by what can be represented in
+ * an unsigned long (for example, 32-bit implementations cannot support a 4GB
+ * size).
+ */
+static void settlbcam(int index, unsigned long virt, phys_addr_t phys,
+		unsigned long size, unsigned long flags, unsigned int pid)
+{
+	unsigned int tsize;
+
+	tsize = __ilog2(size) - 10;
+
+#if defined(CONFIG_SMP) || defined(CONFIG_PPC_E500MC)
+	if ((flags & _PAGE_NO_CACHE) == 0)
+		flags |= _PAGE_COHERENT;
+#endif
+
+	TLBCAM[index].MAS0 = MAS0_TLBSEL(1) | MAS0_ESEL(index) | MAS0_NV(index+1);
+	TLBCAM[index].MAS1 = MAS1_VALID | MAS1_IPROT | MAS1_TSIZE(tsize) | MAS1_TID(pid);
+	TLBCAM[index].MAS2 = virt & PAGE_MASK;
+
+	TLBCAM[index].MAS2 |= (flags & _PAGE_WRITETHRU) ? MAS2_W : 0;
+	TLBCAM[index].MAS2 |= (flags & _PAGE_NO_CACHE) ? MAS2_I : 0;
+	TLBCAM[index].MAS2 |= (flags & _PAGE_COHERENT) ? MAS2_M : 0;
+	TLBCAM[index].MAS2 |= (flags & _PAGE_GUARDED) ? MAS2_G : 0;
+	TLBCAM[index].MAS2 |= (flags & _PAGE_ENDIAN) ? MAS2_E : 0;
+
+	TLBCAM[index].MAS3 = (phys & MAS3_RPN) | MAS3_SR;
+	TLBCAM[index].MAS3 |= (flags & _PAGE_WRITE) ? MAS3_SW : 0;
+	if (mmu_has_feature(MMU_FTR_BIG_PHYS))
+		TLBCAM[index].MAS7 = (u64)phys >> 32;
+
+	/* Below is unlikely -- only for large user pages or similar */
+	if (!is_kernel_addr(virt)) {
+		TLBCAM[index].MAS3 |= MAS3_UR;
+		TLBCAM[index].MAS3 |= (flags & _PAGE_EXEC) ? MAS3_UX : 0;
+		TLBCAM[index].MAS3 |= (flags & _PAGE_WRITE) ? MAS3_UW : 0;
+	} else {
+		TLBCAM[index].MAS3 |= (flags & _PAGE_EXEC) ? MAS3_SX : 0;
+	}
+
+	tlbcam_addrs[index].start = virt;
+	tlbcam_addrs[index].limit = virt + size - 1;
+	tlbcam_addrs[index].phys = phys;
+}
+
+static unsigned long calc_cam_sz(unsigned long ram, unsigned long virt,
+				 phys_addr_t phys)
+{
+	unsigned int camsize = __ilog2(ram);
+	unsigned int align = __ffs(virt | phys);
+	unsigned long max_cam;
+
+	if ((mfspr(SPRN_MMUCFG) & MMUCFG_MAVN) == MMUCFG_MAVN_V1) {
+		/* Convert (4^max) kB to (2^max) bytes */
+		max_cam = ((mfspr(SPRN_TLB1CFG) >> 16) & 0xf) * 2 + 10;
+		camsize &= ~1U;
+		align &= ~1U;
+	} else {
+		/* Convert (2^max) kB to (2^max) bytes */
+		max_cam = __ilog2(mfspr(SPRN_TLB1PS)) + 10;
+	}
+
+	if (camsize > align)
+		camsize = align;
+	if (camsize > max_cam)
+		camsize = max_cam;
+
+	return 1UL << camsize;
+}
+
+static unsigned long map_mem_in_cams_addr(phys_addr_t phys, unsigned long virt,
+					unsigned long ram, int max_cam_idx,
+					bool dryrun, bool init)
+{
+	int i;
+	unsigned long amount_mapped = 0;
+	unsigned long boundary;
+
+	if (strict_kernel_rwx_enabled())
+		boundary = (unsigned long)(_sinittext - _stext);
+	else
+		boundary = ram;
+
+	/* Calculate CAM values */
+	for (i = 0; boundary && i < max_cam_idx; i++) {
+		unsigned long cam_sz;
+		pgprot_t prot = init ? PAGE_KERNEL_X : PAGE_KERNEL_ROX;
+
+		cam_sz = calc_cam_sz(boundary, virt, phys);
+		if (!dryrun)
+			settlbcam(i, virt, phys, cam_sz, pgprot_val(prot), 0);
+
+		boundary -= cam_sz;
+		amount_mapped += cam_sz;
+		virt += cam_sz;
+		phys += cam_sz;
+	}
+	for (ram -= amount_mapped; ram && i < max_cam_idx; i++) {
+		unsigned long cam_sz;
+		pgprot_t prot = init ? PAGE_KERNEL_X : PAGE_KERNEL;
+
+		cam_sz = calc_cam_sz(ram, virt, phys);
+		if (!dryrun)
+			settlbcam(i, virt, phys, cam_sz, pgprot_val(prot), 0);
+
+		ram -= cam_sz;
+		amount_mapped += cam_sz;
+		virt += cam_sz;
+		phys += cam_sz;
+	}
+
+	if (dryrun)
+		return amount_mapped;
+
+	if (init) {
+		loadcam_multi(0, i, max_cam_idx);
+		tlbcam_index = i;
+	} else {
+		loadcam_multi(0, i, 0);
+		WARN_ON(i > tlbcam_index);
+	}
+
+#ifdef CONFIG_PPC64
+	get_paca()->tcd.esel_next = i;
+	get_paca()->tcd.esel_max = mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY;
+	get_paca()->tcd.esel_first = i;
+#endif
+
+	return amount_mapped;
+}
+
+unsigned long map_mem_in_cams(unsigned long ram, int max_cam_idx, bool dryrun, bool init)
+{
+	unsigned long virt = PAGE_OFFSET;
+	phys_addr_t phys = memstart_addr;
+
+	return map_mem_in_cams_addr(phys, virt, ram, max_cam_idx, dryrun, init);
+}
+
+#ifdef CONFIG_PPC32
+
+#if defined(CONFIG_LOWMEM_CAM_NUM_BOOL) && (CONFIG_LOWMEM_CAM_NUM >= NUM_TLBCAMS)
+#error "LOWMEM_CAM_NUM must be less than NUM_TLBCAMS"
+#endif
+
+unsigned long __init mmu_mapin_ram(unsigned long base, unsigned long top)
+{
+	return tlbcam_addrs[tlbcam_index - 1].limit - PAGE_OFFSET + 1;
+}
+
+void flush_instruction_cache(void)
+{
+	unsigned long tmp;
+
+	tmp = mfspr(SPRN_L1CSR1);
+	tmp |= L1CSR1_ICFI | L1CSR1_ICLFR;
+	mtspr(SPRN_L1CSR1, tmp);
+	isync();
+}
+
+/*
+ * MMU_init_hw does the chip-specific initialization of the MMU hardware.
+ */
+void __init MMU_init_hw(void)
+{
+	flush_instruction_cache();
+}
+
+static unsigned long __init tlbcam_sz(int idx)
+{
+	return tlbcam_addrs[idx].limit - tlbcam_addrs[idx].start + 1;
+}
+
+void __init adjust_total_lowmem(void)
+{
+	unsigned long ram;
+	int i;
+
+	/* adjust lowmem size to __max_low_memory */
+	ram = min((phys_addr_t)__max_low_memory, (phys_addr_t)total_lowmem);
+
+	i = switch_to_as1();
+	__max_low_memory = map_mem_in_cams(ram, CONFIG_LOWMEM_CAM_NUM, false, true);
+	restore_to_as0(i, 0, NULL, 1);
+
+	pr_info("Memory CAM mapping: ");
+	for (i = 0; i < tlbcam_index - 1; i++)
+		pr_cont("%lu/", tlbcam_sz(i) >> 20);
+	pr_cont("%lu Mb, residual: %dMb\n", tlbcam_sz(tlbcam_index - 1) >> 20,
+	        (unsigned int)((total_lowmem - __max_low_memory) >> 20));
+
+	memblock_set_current_limit(memstart_addr + __max_low_memory);
+}
+
+#ifdef CONFIG_STRICT_KERNEL_RWX
+int mmu_mark_rodata_ro(void)
+{
+	unsigned long remapped;
+
+	remapped = map_mem_in_cams(__max_low_memory, CONFIG_LOWMEM_CAM_NUM, false, false);
+
+	if (WARN_ON(__max_low_memory != remapped))
+		return -EINVAL;
+
+	return 0;
+}
+#endif
+
+int mmu_mark_initmem_nx(void)
+{
+	/* Everything is done in mmu_mark_rodata_ro() */
+	return 0;
+}
+
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	phys_addr_t limit = first_memblock_base + first_memblock_size;
+
+	/* 64M mapped initially according to head_fsl_booke.S */
+	memblock_set_current_limit(min_t(u64, limit, 0x04000000));
+}
+
+#ifdef CONFIG_RELOCATABLE
+int __initdata is_second_reloc;
+notrace void __init relocate_init(u64 dt_ptr, phys_addr_t start)
+{
+	unsigned long base = kernstart_virt_addr;
+	phys_addr_t size;
+
+	kernstart_addr = start;
+	if (is_second_reloc) {
+		virt_phys_offset = PAGE_OFFSET - memstart_addr;
+		kaslr_late_init();
+		return;
+	}
+
+	/*
+	 * Relocatable kernel support based on processing of dynamic
+	 * relocation entries. Before we get the real memstart_addr,
+	 * We will compute the virt_phys_offset like this:
+	 * virt_phys_offset = stext.run - kernstart_addr
+	 *
+	 * stext.run = (KERNELBASE & ~0x3ffffff) +
+	 *				(kernstart_addr & 0x3ffffff)
+	 * When we relocate, we have :
+	 *
+	 *	(kernstart_addr & 0x3ffffff) = (stext.run & 0x3ffffff)
+	 *
+	 * hence:
+	 *  virt_phys_offset = (KERNELBASE & ~0x3ffffff) -
+	 *                              (kernstart_addr & ~0x3ffffff)
+	 *
+	 */
+	start &= ~0x3ffffff;
+	base &= ~0x3ffffff;
+	virt_phys_offset = base - start;
+	early_get_first_memblock_info(__va(dt_ptr), &size);
+	/*
+	 * We now get the memstart_addr, then we should check if this
+	 * address is the same as what the PAGE_OFFSET map to now. If
+	 * not we have to change the map of PAGE_OFFSET to memstart_addr
+	 * and do a second relocation.
+	 */
+	if (start != memstart_addr) {
+		int n;
+		long offset = start - memstart_addr;
+
+		is_second_reloc = 1;
+		n = switch_to_as1();
+		/* map a 64M area for the second relocation */
+		if (memstart_addr > start)
+			map_mem_in_cams(0x4000000, CONFIG_LOWMEM_CAM_NUM,
+					false, true);
+		else
+			map_mem_in_cams_addr(start, PAGE_OFFSET + offset,
+					0x4000000, CONFIG_LOWMEM_CAM_NUM,
+					false, true);
+		restore_to_as0(n, offset, __va(dt_ptr), 1);
+		/* We should never reach here */
+		panic("Relocation error");
+	}
+
+	kaslr_early_init(__va(dt_ptr), size);
+}
+#endif
+#endif

linux-7.2.3/arch/powerpc/mm/nohash/e500_hugetlbpage.c

@@ -0,0 +1,193 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * PPC Huge TLB Page Support for Book3E MMU
+ *
+ * Copyright (C) 2009 David Gibson, IBM Corporation.
+ * Copyright (C) 2011 Becky Bruce, Freescale Semiconductor
+ *
+ */
+#include <linux/mm.h>
+#include <linux/hugetlb.h>
+
+#include <asm/mmu.h>
+
+#ifdef CONFIG_PPC64
+#include <asm/paca.h>
+
+static inline int tlb1_next(void)
+{
+	struct paca_struct *paca = get_paca();
+	struct tlb_core_data *tcd;
+	int this, next;
+
+	tcd = paca->tcd_ptr;
+	this = tcd->esel_next;
+
+	next = this + 1;
+	if (next >= tcd->esel_max)
+		next = tcd->esel_first;
+
+	tcd->esel_next = next;
+	return this;
+}
+
+static inline void book3e_tlb_lock(void)
+{
+	struct paca_struct *paca = get_paca();
+	unsigned long tmp;
+	int token = smp_processor_id() + 1;
+
+	/*
+	 * Besides being unnecessary in the absence of SMT, this
+	 * check prevents trying to do lbarx/stbcx. on e5500 which
+	 * doesn't implement either feature.
+	 */
+	if (!cpu_has_feature(CPU_FTR_SMT))
+		return;
+
+	asm volatile(".machine push;"
+		     ".machine e6500;"
+		     "1: lbarx %0, 0, %1;"
+		     "cmpwi %0, 0;"
+		     "bne 2f;"
+		     "stbcx. %2, 0, %1;"
+		     "bne 1b;"
+		     "b 3f;"
+		     "2: lbzx %0, 0, %1;"
+		     "cmpwi %0, 0;"
+		     "bne 2b;"
+		     "b 1b;"
+		     "3:"
+		     ".machine pop;"
+		     : "=&r" (tmp)
+		     : "r" (&paca->tcd_ptr->lock), "r" (token)
+		     : "memory");
+}
+
+static inline void book3e_tlb_unlock(void)
+{
+	struct paca_struct *paca = get_paca();
+
+	if (!cpu_has_feature(CPU_FTR_SMT))
+		return;
+
+	isync();
+	paca->tcd_ptr->lock = 0;
+}
+#else
+static inline int tlb1_next(void)
+{
+	int index, ncams;
+
+	ncams = mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY;
+
+	index = this_cpu_read(next_tlbcam_idx);
+
+	/* Just round-robin the entries and wrap when we hit the end */
+	if (unlikely(index == ncams - 1))
+		__this_cpu_write(next_tlbcam_idx, tlbcam_index);
+	else
+		__this_cpu_inc(next_tlbcam_idx);
+
+	return index;
+}
+
+static inline void book3e_tlb_lock(void)
+{
+}
+
+static inline void book3e_tlb_unlock(void)
+{
+}
+#endif
+
+static inline int book3e_tlb_exists(unsigned long ea, unsigned long pid)
+{
+	int found = 0;
+
+	mtspr(SPRN_MAS6, pid << 16);
+	asm volatile(
+		"tlbsx	0,%1\n"
+		"mfspr	%0,0x271\n"
+		"srwi	%0,%0,31\n"
+		: "=&r"(found) : "r"(ea));
+
+	return found;
+}
+
+static void
+book3e_hugetlb_preload(struct vm_area_struct *vma, unsigned long ea, pte_t pte)
+{
+	unsigned long mas1, mas2;
+	u64 mas7_3;
+	unsigned long psize, tsize, shift;
+	unsigned long flags;
+	struct mm_struct *mm;
+	int index;
+
+	if (unlikely(is_kernel_addr(ea)))
+		return;
+
+	mm = vma->vm_mm;
+
+	psize = vma_mmu_pagesize(vma);
+	shift = __ilog2(psize);
+	tsize = shift - 10;
+	/*
+	 * We can't be interrupted while we're setting up the MAS
+	 * registers or after we've confirmed that no tlb exists.
+	 */
+	local_irq_save(flags);
+
+	book3e_tlb_lock();
+
+	if (unlikely(book3e_tlb_exists(ea, mm->context.id))) {
+		book3e_tlb_unlock();
+		local_irq_restore(flags);
+		return;
+	}
+
+	/* We have to use the CAM(TLB1) on FSL parts for hugepages */
+	index = tlb1_next();
+	mtspr(SPRN_MAS0, MAS0_ESEL(index) | MAS0_TLBSEL(1));
+
+	mas1 = MAS1_VALID | MAS1_TID(mm->context.id) | MAS1_TSIZE(tsize);
+	mas2 = ea & ~((1UL << shift) - 1);
+	mas2 |= (pte_val(pte) >> PTE_WIMGE_SHIFT) & MAS2_WIMGE_MASK;
+	mas7_3 = (u64)pte_pfn(pte) << PAGE_SHIFT;
+	mas7_3 |= (pte_val(pte) >> PTE_BAP_SHIFT) & MAS3_BAP_MASK;
+	if (!pte_dirty(pte))
+		mas7_3 &= ~(MAS3_SW|MAS3_UW);
+
+	mtspr(SPRN_MAS1, mas1);
+	mtspr(SPRN_MAS2, mas2);
+
+	if (mmu_has_feature(MMU_FTR_BIG_PHYS))
+		mtspr(SPRN_MAS7, upper_32_bits(mas7_3));
+	mtspr(SPRN_MAS3, lower_32_bits(mas7_3));
+
+	asm volatile ("tlbwe");
+
+	book3e_tlb_unlock();
+	local_irq_restore(flags);
+}
+
+/*
+ * This is called at the end of handling a user page fault, when the
+ * fault has been handled by updating a PTE in the linux page tables.
+ *
+ * This must always be called with the pte lock held.
+ */
+void __update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t *ptep)
+{
+	if (is_vm_hugetlb_page(vma))
+		book3e_hugetlb_preload(vma, address, *ptep);
+}
+
+void flush_hugetlb_page(struct vm_area_struct *vma, unsigned long vmaddr)
+{
+	struct hstate *hstate = hstate_file(vma->vm_file);
+	unsigned long tsize = huge_page_shift(hstate) - 10;
+
+	__flush_tlb_page(vma->vm_mm, vmaddr, tsize, 0);
+}

linux-7.2.3/arch/powerpc/mm/nohash/kaslr_booke.c

@@ -0,0 +1,395 @@
+// SPDX-License-Identifier: GPL-2.0-only
+//
+// Copyright (C) 2019 Jason Yan <yanaijie@huawei.com>
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/stddef.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/memblock.h>
+#include <linux/libfdt.h>
+#include <linux/crash_reserve.h>
+#include <linux/of.h>
+#include <linux/of_fdt.h>
+#include <asm/cacheflush.h>
+#include <asm/kdump.h>
+#include <mm/mmu_decl.h>
+
+struct regions {
+	unsigned long pa_start;
+	unsigned long pa_end;
+	unsigned long kernel_size;
+	unsigned long dtb_start;
+	unsigned long dtb_end;
+	unsigned long initrd_start;
+	unsigned long initrd_end;
+	unsigned long crash_start;
+	unsigned long crash_end;
+	int reserved_mem;
+	int reserved_mem_addr_cells;
+	int reserved_mem_size_cells;
+};
+
+struct regions __initdata regions;
+
+static __init void kaslr_get_cmdline(void *fdt)
+{
+	early_init_dt_scan_chosen(boot_command_line);
+}
+
+static unsigned long __init rotate_xor(unsigned long hash, const void *area,
+				       size_t size)
+{
+	size_t i;
+	const unsigned long *ptr = area;
+
+	for (i = 0; i < size / sizeof(hash); i++) {
+		/* Rotate by odd number of bits and XOR. */
+		hash = (hash << ((sizeof(hash) * 8) - 7)) | (hash >> 7);
+		hash ^= ptr[i];
+	}
+
+	return hash;
+}
+
+/* Attempt to create a simple starting entropy. This can make it defferent for
+ * every build but it is still not enough. Stronger entropy should
+ * be added to make it change for every boot.
+ */
+static unsigned long __init get_boot_seed(void *fdt)
+{
+	unsigned long hash = 0;
+
+	/* build-specific string for starting entropy. */
+	hash = rotate_xor(hash, linux_banner, strlen(linux_banner));
+	hash = rotate_xor(hash, fdt, fdt_totalsize(fdt));
+
+	return hash;
+}
+
+static __init u64 get_kaslr_seed(void *fdt)
+{
+	int node, len;
+	fdt64_t *prop;
+	u64 ret;
+
+	node = fdt_path_offset(fdt, "/chosen");
+	if (node < 0)
+		return 0;
+
+	prop = fdt_getprop_w(fdt, node, "kaslr-seed", &len);
+	if (!prop || len != sizeof(u64))
+		return 0;
+
+	ret = fdt64_to_cpu(*prop);
+	*prop = 0;
+	return ret;
+}
+
+static __init bool regions_overlap(u32 s1, u32 e1, u32 s2, u32 e2)
+{
+	return e1 >= s2 && e2 >= s1;
+}
+
+static __init bool overlaps_reserved_region(const void *fdt, u32 start,
+					    u32 end)
+{
+	int subnode, len, i;
+	u64 base, size;
+
+	/* check for overlap with /memreserve/ entries */
+	for (i = 0; i < fdt_num_mem_rsv(fdt); i++) {
+		if (fdt_get_mem_rsv(fdt, i, &base, &size) < 0)
+			continue;
+		if (regions_overlap(start, end, base, base + size))
+			return true;
+	}
+
+	if (regions.reserved_mem < 0)
+		return false;
+
+	/* check for overlap with static reservations in /reserved-memory */
+	for (subnode = fdt_first_subnode(fdt, regions.reserved_mem);
+	     subnode >= 0;
+	     subnode = fdt_next_subnode(fdt, subnode)) {
+		const fdt32_t *reg;
+		u64 rsv_end;
+
+		len = 0;
+		reg = fdt_getprop(fdt, subnode, "reg", &len);
+		while (len >= (regions.reserved_mem_addr_cells +
+			       regions.reserved_mem_size_cells)) {
+			base = fdt32_to_cpu(reg[0]);
+			if (regions.reserved_mem_addr_cells == 2)
+				base = (base << 32) | fdt32_to_cpu(reg[1]);
+
+			reg += regions.reserved_mem_addr_cells;
+			len -= 4 * regions.reserved_mem_addr_cells;
+
+			size = fdt32_to_cpu(reg[0]);
+			if (regions.reserved_mem_size_cells == 2)
+				size = (size << 32) | fdt32_to_cpu(reg[1]);
+
+			reg += regions.reserved_mem_size_cells;
+			len -= 4 * regions.reserved_mem_size_cells;
+
+			if (base >= regions.pa_end)
+				continue;
+
+			rsv_end = min(base + size, (u64)U32_MAX);
+
+			if (regions_overlap(start, end, base, rsv_end))
+				return true;
+		}
+	}
+	return false;
+}
+
+static __init bool overlaps_region(const void *fdt, u32 start,
+				   u32 end)
+{
+	if (regions_overlap(start, end, __pa(_stext), __pa(_end)))
+		return true;
+
+	if (regions_overlap(start, end, regions.dtb_start,
+			    regions.dtb_end))
+		return true;
+
+	if (regions_overlap(start, end, regions.initrd_start,
+			    regions.initrd_end))
+		return true;
+
+	if (regions_overlap(start, end, regions.crash_start,
+			    regions.crash_end))
+		return true;
+
+	return overlaps_reserved_region(fdt, start, end);
+}
+
+static void __init get_crash_kernel(void *fdt, unsigned long size)
+{
+#ifdef CONFIG_CRASH_RESERVE
+	unsigned long long crash_size, crash_base;
+	int ret;
+
+	ret = parse_crashkernel(boot_command_line, size, &crash_size,
+				&crash_base, NULL, NULL, NULL);
+	if (ret != 0 || crash_size == 0)
+		return;
+	if (crash_base == 0)
+		crash_base = KDUMP_KERNELBASE;
+
+	regions.crash_start = (unsigned long)crash_base;
+	regions.crash_end = (unsigned long)(crash_base + crash_size);
+
+	pr_debug("crash_base=0x%llx crash_size=0x%llx\n", crash_base, crash_size);
+#endif
+}
+
+static void __init get_initrd_range(void *fdt)
+{
+	u64 start, end;
+	int node, len;
+	const __be32 *prop;
+
+	node = fdt_path_offset(fdt, "/chosen");
+	if (node < 0)
+		return;
+
+	prop = fdt_getprop(fdt, node, "linux,initrd-start", &len);
+	if (!prop)
+		return;
+	start = of_read_number(prop, len / 4);
+
+	prop = fdt_getprop(fdt, node, "linux,initrd-end", &len);
+	if (!prop)
+		return;
+	end = of_read_number(prop, len / 4);
+
+	regions.initrd_start = (unsigned long)start;
+	regions.initrd_end = (unsigned long)end;
+
+	pr_debug("initrd_start=0x%llx  initrd_end=0x%llx\n", start, end);
+}
+
+static __init unsigned long get_usable_address(const void *fdt,
+					       unsigned long start,
+					       unsigned long offset)
+{
+	unsigned long pa;
+	unsigned long pa_end;
+
+	for (pa = offset; (long)pa > (long)start; pa -= SZ_16K) {
+		pa_end = pa + regions.kernel_size;
+		if (overlaps_region(fdt, pa, pa_end))
+			continue;
+
+		return pa;
+	}
+	return 0;
+}
+
+static __init void get_cell_sizes(const void *fdt, int node, int *addr_cells,
+				  int *size_cells)
+{
+	const int *prop;
+	int len;
+
+	/*
+	 * Retrieve the #address-cells and #size-cells properties
+	 * from the 'node', or use the default if not provided.
+	 */
+	*addr_cells = *size_cells = 1;
+
+	prop = fdt_getprop(fdt, node, "#address-cells", &len);
+	if (len == 4)
+		*addr_cells = fdt32_to_cpu(*prop);
+	prop = fdt_getprop(fdt, node, "#size-cells", &len);
+	if (len == 4)
+		*size_cells = fdt32_to_cpu(*prop);
+}
+
+static unsigned long __init kaslr_legal_offset(void *dt_ptr, unsigned long index,
+					       unsigned long offset)
+{
+	unsigned long koffset = 0;
+	unsigned long start;
+
+	while ((long)index >= 0) {
+		offset = memstart_addr + index * SZ_64M + offset;
+		start = memstart_addr + index * SZ_64M;
+		koffset = get_usable_address(dt_ptr, start, offset);
+		if (koffset)
+			break;
+		index--;
+	}
+
+	if (koffset != 0)
+		koffset -= memstart_addr;
+
+	return koffset;
+}
+
+static inline __init bool kaslr_disabled(void)
+{
+	return strstr(boot_command_line, "nokaslr") != NULL;
+}
+
+static unsigned long __init kaslr_choose_location(void *dt_ptr, phys_addr_t size,
+						  unsigned long kernel_sz)
+{
+	unsigned long offset, random;
+	unsigned long ram, linear_sz;
+	u64 seed;
+	unsigned long index;
+
+	kaslr_get_cmdline(dt_ptr);
+	if (kaslr_disabled())
+		return 0;
+
+	random = get_boot_seed(dt_ptr);
+
+	seed = get_tb() << 32;
+	seed ^= get_tb();
+	random = rotate_xor(random, &seed, sizeof(seed));
+
+	/*
+	 * Retrieve (and wipe) the seed from the FDT
+	 */
+	seed = get_kaslr_seed(dt_ptr);
+	if (seed)
+		random = rotate_xor(random, &seed, sizeof(seed));
+	else
+		pr_warn("KASLR: No safe seed for randomizing the kernel base.\n");
+
+	ram = min_t(phys_addr_t, __max_low_memory, size);
+	ram = map_mem_in_cams(ram, CONFIG_LOWMEM_CAM_NUM, true, true);
+	linear_sz = min_t(unsigned long, ram, SZ_512M);
+
+	/* If the linear size is smaller than 64M, do not randomize */
+	if (linear_sz < SZ_64M)
+		return 0;
+
+	/* check for a reserved-memory node and record its cell sizes */
+	regions.reserved_mem = fdt_path_offset(dt_ptr, "/reserved-memory");
+	if (regions.reserved_mem >= 0)
+		get_cell_sizes(dt_ptr, regions.reserved_mem,
+			       &regions.reserved_mem_addr_cells,
+			       &regions.reserved_mem_size_cells);
+
+	regions.pa_start = memstart_addr;
+	regions.pa_end = memstart_addr + linear_sz;
+	regions.dtb_start = __pa(dt_ptr);
+	regions.dtb_end = __pa(dt_ptr) + fdt_totalsize(dt_ptr);
+	regions.kernel_size = kernel_sz;
+
+	get_initrd_range(dt_ptr);
+	get_crash_kernel(dt_ptr, ram);
+
+	/*
+	 * Decide which 64M we want to start
+	 * Only use the low 8 bits of the random seed
+	 */
+	index = random & 0xFF;
+	index %= linear_sz / SZ_64M;
+
+	/* Decide offset inside 64M */
+	offset = random % (SZ_64M - kernel_sz);
+	offset = round_down(offset, SZ_16K);
+
+	return kaslr_legal_offset(dt_ptr, index, offset);
+}
+
+/*
+ * To see if we need to relocate the kernel to a random offset
+ * void *dt_ptr - address of the device tree
+ * phys_addr_t size - size of the first memory block
+ */
+notrace void __init kaslr_early_init(void *dt_ptr, phys_addr_t size)
+{
+	unsigned long tlb_virt;
+	phys_addr_t tlb_phys;
+	unsigned long offset;
+	unsigned long kernel_sz;
+
+	kernel_sz = (unsigned long)_end - (unsigned long)_stext;
+
+	offset = kaslr_choose_location(dt_ptr, size, kernel_sz);
+	if (offset == 0)
+		return;
+
+	kernstart_virt_addr += offset;
+	kernstart_addr += offset;
+
+	is_second_reloc = 1;
+
+	if (offset >= SZ_64M) {
+		tlb_virt = round_down(kernstart_virt_addr, SZ_64M);
+		tlb_phys = round_down(kernstart_addr, SZ_64M);
+
+		/* Create kernel map to relocate in */
+		create_kaslr_tlb_entry(1, tlb_virt, tlb_phys);
+	}
+
+	/* Copy the kernel to its new location and run */
+	memcpy((void *)kernstart_virt_addr, (void *)_stext, kernel_sz);
+	flush_icache_range(kernstart_virt_addr, kernstart_virt_addr + kernel_sz);
+
+	reloc_kernel_entry(dt_ptr, kernstart_virt_addr);
+}
+
+void __init kaslr_late_init(void)
+{
+	/* If randomized, clear the original kernel */
+	if (kernstart_virt_addr != KERNELBASE) {
+		unsigned long kernel_sz;
+
+		kernel_sz = (unsigned long)_end - kernstart_virt_addr;
+		memzero_explicit((void *)KERNELBASE, kernel_sz);
+	}
+}

linux-7.2.3/arch/powerpc/mm/nohash/kup.c

@@ -0,0 +1,27 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * This file contains the routines for initializing kernel userspace protection
+ */
+
+#include <linux/export.h>
+#include <linux/init.h>
+#include <linux/printk.h>
+#include <linux/smp.h>
+
+#include <asm/kup.h>
+#include <asm/smp.h>
+
+#ifdef CONFIG_PPC_KUAP
+void setup_kuap(bool disabled)
+{
+	if (disabled) {
+		if (smp_processor_id() == boot_cpuid)
+			cur_cpu_spec->mmu_features &= ~MMU_FTR_KUAP;
+		return;
+	}
+
+	pr_info("Activating Kernel Userspace Access Protection\n");
+
+	prevent_user_access(KUAP_READ_WRITE);
+}
+#endif

linux-7.2.3/arch/powerpc/mm/nohash/mmu_context.c

@@ -0,0 +1,404 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * This file contains the routines for handling the MMU on those
+ * PowerPC implementations where the MMU is not using the hash
+ * table, such as 8xx, 4xx, BookE's etc...
+ *
+ * Copyright 2008 Ben Herrenschmidt <benh@kernel.crashing.org>
+ *                IBM Corp.
+ *
+ *  Derived from previous arch/powerpc/mm/mmu_context.c
+ *  and arch/powerpc/include/asm/mmu_context.h
+ *
+ * TODO:
+ *
+ *   - The global context lock will not scale very well
+ *   - The maps should be dynamically allocated to allow for processors
+ *     that support more PID bits at runtime
+ *   - Implement flush_tlb_mm() by making the context stale and picking
+ *     a new one
+ *   - More aggressively clear stale map bits and maybe find some way to
+ *     also clear mm->cpu_vm_mask bits when processes are migrated
+ */
+
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/memblock.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
+#include <linux/slab.h>
+
+#include <asm/mmu_context.h>
+#include <asm/tlbflush.h>
+#include <asm/smp.h>
+#include <asm/kup.h>
+
+#include <mm/mmu_decl.h>
+
+/*
+ * Room for two PTE table pointers, usually the kernel and current user
+ * pointer to their respective root page table (pgdir).
+ */
+void *abatron_pteptrs[2];
+
+/*
+ * The MPC8xx has only 16 contexts. We rotate through them on each task switch.
+ * A better way would be to keep track of tasks that own contexts, and implement
+ * an LRU usage. That way very active tasks don't always have to pay the TLB
+ * reload overhead. The kernel pages are mapped shared, so the kernel can run on
+ * behalf of any task that makes a kernel entry. Shared does not mean they are
+ * not protected, just that the ASID comparison is not performed. -- Dan
+ *
+ * The IBM4xx has 256 contexts, so we can just rotate through these as a way of
+ * "switching" contexts. If the TID of the TLB is zero, the PID/TID comparison
+ * is disabled, so we can use a TID of zero to represent all kernel pages as
+ * shared among all contexts. -- Dan
+ *
+ * The IBM 47x core supports 16-bit PIDs, thus 65535 contexts. We should
+ * normally never have to steal though the facility is present if needed.
+ * -- BenH
+ */
+#define FIRST_CONTEXT 1
+#if defined(CONFIG_PPC_8xx)
+#define LAST_CONTEXT 16
+#elif defined(CONFIG_PPC_47x)
+#define LAST_CONTEXT 65535
+#else
+#define LAST_CONTEXT 255
+#endif
+
+static unsigned int next_context, nr_free_contexts;
+static unsigned long *context_map;
+static unsigned long *stale_map[NR_CPUS];
+static struct mm_struct **context_mm;
+static DEFINE_RAW_SPINLOCK(context_lock);
+
+#define CTX_MAP_SIZE	\
+	(sizeof(unsigned long) * (LAST_CONTEXT / BITS_PER_LONG + 1))
+
+
+/* Steal a context from a task that has one at the moment.
+ *
+ * This is used when we are running out of available PID numbers
+ * on the processors.
+ *
+ * This isn't an LRU system, it just frees up each context in
+ * turn (sort-of pseudo-random replacement :).  This would be the
+ * place to implement an LRU scheme if anyone was motivated to do it.
+ *  -- paulus
+ *
+ * For context stealing, we use a slightly different approach for
+ * SMP and UP. Basically, the UP one is simpler and doesn't use
+ * the stale map as we can just flush the local CPU
+ *  -- benh
+ */
+static unsigned int steal_context_smp(unsigned int id)
+{
+	struct mm_struct *mm;
+	unsigned int cpu, max, i;
+
+	max = LAST_CONTEXT - FIRST_CONTEXT;
+
+	/* Attempt to free next_context first and then loop until we manage */
+	while (max--) {
+		/* Pick up the victim mm */
+		mm = context_mm[id];
+
+		/* We have a candidate victim, check if it's active, on SMP
+		 * we cannot steal active contexts
+		 */
+		if (mm->context.active) {
+			id++;
+			if (id > LAST_CONTEXT)
+				id = FIRST_CONTEXT;
+			continue;
+		}
+
+		/* Mark this mm has having no context anymore */
+		mm->context.id = MMU_NO_CONTEXT;
+
+		/* Mark it stale on all CPUs that used this mm. For threaded
+		 * implementations, we set it on all threads on each core
+		 * represented in the mask. A future implementation will use
+		 * a core map instead but this will do for now.
+		 */
+		for_each_cpu(cpu, mm_cpumask(mm)) {
+			for (i = cpu_first_thread_sibling(cpu);
+			     i <= cpu_last_thread_sibling(cpu); i++) {
+				if (stale_map[i])
+					__set_bit(id, stale_map[i]);
+			}
+			cpu = i - 1;
+		}
+		return id;
+	}
+
+	/* This will happen if you have more CPUs than available contexts,
+	 * all we can do here is wait a bit and try again
+	 */
+	raw_spin_unlock(&context_lock);
+	cpu_relax();
+	raw_spin_lock(&context_lock);
+
+	/* This will cause the caller to try again */
+	return MMU_NO_CONTEXT;
+}
+
+static unsigned int steal_all_contexts(void)
+{
+	struct mm_struct *mm;
+	int cpu = smp_processor_id();
+	unsigned int id;
+
+	for (id = FIRST_CONTEXT; id <= LAST_CONTEXT; id++) {
+		/* Pick up the victim mm */
+		mm = context_mm[id];
+
+		/* Mark this mm as having no context anymore */
+		mm->context.id = MMU_NO_CONTEXT;
+		if (id != FIRST_CONTEXT) {
+			context_mm[id] = NULL;
+			__clear_bit(id, context_map);
+		}
+		if (IS_ENABLED(CONFIG_SMP))
+			__clear_bit(id, stale_map[cpu]);
+	}
+
+	/* Flush the TLB for all contexts (not to be used on SMP) */
+	_tlbil_all();
+
+	nr_free_contexts = LAST_CONTEXT - FIRST_CONTEXT;
+
+	return FIRST_CONTEXT;
+}
+
+/* Note that this will also be called on SMP if all other CPUs are
+ * offlined, which means that it may be called for cpu != 0. For
+ * this to work, we somewhat assume that CPUs that are onlined
+ * come up with a fully clean TLB (or are cleaned when offlined)
+ */
+static unsigned int steal_context_up(unsigned int id)
+{
+	struct mm_struct *mm;
+	int cpu = smp_processor_id();
+
+	/* Pick up the victim mm */
+	mm = context_mm[id];
+
+	/* Flush the TLB for that context */
+	local_flush_tlb_mm(mm);
+
+	/* Mark this mm has having no context anymore */
+	mm->context.id = MMU_NO_CONTEXT;
+
+	/* XXX This clear should ultimately be part of local_flush_tlb_mm */
+	if (IS_ENABLED(CONFIG_SMP))
+		__clear_bit(id, stale_map[cpu]);
+
+	return id;
+}
+
+static void set_context(unsigned long id, pgd_t *pgd)
+{
+	if (IS_ENABLED(CONFIG_PPC_8xx)) {
+		mtspr(SPRN_M_TWB, __pa(pgd));
+
+		/* Update context */
+		mtspr(SPRN_M_CASID, id - 1);
+
+		/* sync */
+		mb();
+	} else if (kuap_is_disabled()) {
+		mtspr(SPRN_PID, id);
+		isync();
+	}
+}
+
+void switch_mmu_context(struct mm_struct *prev, struct mm_struct *next,
+			struct task_struct *tsk)
+{
+	unsigned int id;
+	unsigned int i, cpu = smp_processor_id();
+	unsigned long *map;
+
+	/* No lockless fast path .. yet */
+	raw_spin_lock(&context_lock);
+
+	if (IS_ENABLED(CONFIG_SMP)) {
+		/* Mark us active and the previous one not anymore */
+		next->context.active++;
+		if (prev) {
+			WARN_ON(prev->context.active < 1);
+			prev->context.active--;
+		}
+	}
+
+ again:
+
+	/* If we already have a valid assigned context, skip all that */
+	id = next->context.id;
+	if (likely(id != MMU_NO_CONTEXT))
+		goto ctxt_ok;
+
+	/* We really don't have a context, let's try to acquire one */
+	id = next_context;
+	if (id > LAST_CONTEXT)
+		id = FIRST_CONTEXT;
+	map = context_map;
+
+	/* No more free contexts, let's try to steal one */
+	if (nr_free_contexts == 0) {
+		if (num_online_cpus() > 1) {
+			id = steal_context_smp(id);
+			if (id == MMU_NO_CONTEXT)
+				goto again;
+			goto stolen;
+		}
+		if (IS_ENABLED(CONFIG_PPC_8xx))
+			id = steal_all_contexts();
+		else
+			id = steal_context_up(id);
+		goto stolen;
+	}
+	nr_free_contexts--;
+
+	/* We know there's at least one free context, try to find it */
+	while (__test_and_set_bit(id, map)) {
+		id = find_next_zero_bit(map, LAST_CONTEXT+1, id);
+		if (id > LAST_CONTEXT)
+			id = FIRST_CONTEXT;
+	}
+ stolen:
+	next_context = id + 1;
+	context_mm[id] = next;
+	next->context.id = id;
+
+ ctxt_ok:
+
+	/* If that context got marked stale on this CPU, then flush the
+	 * local TLB for it and unmark it before we use it
+	 */
+	if (IS_ENABLED(CONFIG_SMP) && test_bit(id, stale_map[cpu])) {
+		local_flush_tlb_mm(next);
+
+		/* XXX This clear should ultimately be part of local_flush_tlb_mm */
+		for (i = cpu_first_thread_sibling(cpu);
+		     i <= cpu_last_thread_sibling(cpu); i++) {
+			if (stale_map[i])
+				__clear_bit(id, stale_map[i]);
+		}
+	}
+
+	/* Flick the MMU and release lock */
+	if (IS_ENABLED(CONFIG_BDI_SWITCH))
+		abatron_pteptrs[1] = next->pgd;
+	set_context(id, next->pgd);
+#if defined(CONFIG_BOOKE) && defined(CONFIG_PPC_KUAP)
+	tsk->thread.pid = id;
+#endif
+	raw_spin_unlock(&context_lock);
+}
+
+/*
+ * Set up the context for a new address space.
+ */
+int init_new_context(struct task_struct *t, struct mm_struct *mm)
+{
+	mm->context.id = MMU_NO_CONTEXT;
+	mm->context.active = 0;
+	pte_frag_set(&mm->context, NULL);
+	return 0;
+}
+
+/*
+ * We're finished using the context for an address space.
+ */
+void destroy_context(struct mm_struct *mm)
+{
+	unsigned long flags;
+	unsigned int id;
+
+	if (mm->context.id == MMU_NO_CONTEXT)
+		return;
+
+	WARN_ON(mm->context.active != 0);
+
+	raw_spin_lock_irqsave(&context_lock, flags);
+	id = mm->context.id;
+	if (id != MMU_NO_CONTEXT) {
+		__clear_bit(id, context_map);
+		mm->context.id = MMU_NO_CONTEXT;
+		context_mm[id] = NULL;
+		nr_free_contexts++;
+	}
+	raw_spin_unlock_irqrestore(&context_lock, flags);
+}
+
+static int mmu_ctx_cpu_prepare(unsigned int cpu)
+{
+	/* We don't touch CPU 0 map, it's allocated at aboot and kept
+	 * around forever
+	 */
+	if (cpu == boot_cpuid)
+		return 0;
+
+	stale_map[cpu] = kzalloc(CTX_MAP_SIZE, GFP_KERNEL);
+	return 0;
+}
+
+static int mmu_ctx_cpu_dead(unsigned int cpu)
+{
+#ifdef CONFIG_HOTPLUG_CPU
+	if (cpu == boot_cpuid)
+		return 0;
+
+	kfree(stale_map[cpu]);
+	stale_map[cpu] = NULL;
+
+	/* We also clear the cpu_vm_mask bits of CPUs going away */
+	clear_tasks_mm_cpumask(cpu);
+#endif
+	return 0;
+}
+
+/*
+ * Initialize the context management stuff.
+ */
+void __init mmu_context_init(void)
+{
+	/* Mark init_mm as being active on all possible CPUs since
+	 * we'll get called with prev == init_mm the first time
+	 * we schedule on a given CPU
+	 */
+	init_mm.context.active = NR_CPUS;
+
+	/*
+	 * Allocate the maps used by context management
+	 */
+	context_map = memblock_alloc_or_panic(CTX_MAP_SIZE, SMP_CACHE_BYTES);
+	context_mm = memblock_alloc_or_panic(sizeof(void *) * (LAST_CONTEXT + 1),
+				    SMP_CACHE_BYTES);
+	if (IS_ENABLED(CONFIG_SMP)) {
+		stale_map[boot_cpuid] = memblock_alloc_or_panic(CTX_MAP_SIZE, SMP_CACHE_BYTES);
+		cpuhp_setup_state_nocalls(CPUHP_POWERPC_MMU_CTX_PREPARE,
+					  "powerpc/mmu/ctx:prepare",
+					  mmu_ctx_cpu_prepare, mmu_ctx_cpu_dead);
+	}
+
+	printk(KERN_INFO
+	       "MMU: Allocated %zu bytes of context maps for %d contexts\n",
+	       2 * CTX_MAP_SIZE + (sizeof(void *) * (LAST_CONTEXT + 1)),
+	       LAST_CONTEXT - FIRST_CONTEXT + 1);
+
+	/*
+	 * Some processors have too few contexts to reserve one for
+	 * init_mm, and require using context 0 for a normal task.
+	 * Other processors reserve the use of context zero for the kernel.
+	 * This code assumes FIRST_CONTEXT < 32.
+	 */
+	context_map[0] = (1 << FIRST_CONTEXT) - 1;
+	next_context = FIRST_CONTEXT;
+	nr_free_contexts = LAST_CONTEXT - FIRST_CONTEXT + 1;
+}

linux-7.2.3/arch/powerpc/mm/nohash/tlb.c

@@ -0,0 +1,341 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * This file contains the routines for TLB flushing.
+ * On machines where the MMU does not use a hash table to store virtual to
+ * physical translations (ie, SW loaded TLBs or Book3E compilant processors,
+ * this does -not- include 603 however which shares the implementation with
+ * hash based processors)
+ *
+ *  -- BenH
+ *
+ * Copyright 2008,2009 Ben Herrenschmidt <benh@kernel.crashing.org>
+ *                     IBM Corp.
+ *
+ *  Derived from arch/ppc/mm/init.c:
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au)
+ *  and Cort Dougan (PReP) (cort@cs.nmt.edu)
+ *    Copyright (C) 1996 Paul Mackerras
+ *
+ *  Derived from "arch/i386/mm/init.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ */
+
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/highmem.h>
+#include <linux/pagemap.h>
+#include <linux/preempt.h>
+#include <linux/spinlock.h>
+#include <linux/memblock.h>
+#include <linux/of_fdt.h>
+#include <linux/hugetlb.h>
+
+#include <asm/pgalloc.h>
+#include <asm/tlbflush.h>
+#include <asm/tlb.h>
+#include <asm/text-patching.h>
+#include <asm/cputhreads.h>
+#include <asm/hugetlb.h>
+#include <asm/paca.h>
+
+#include <mm/mmu_decl.h>
+
+/*
+ * This struct lists the sw-supported page sizes.  The hardawre MMU may support
+ * other sizes not listed here.   The .ind field is only used on MMUs that have
+ * indirect page table entries.
+ */
+#ifdef CONFIG_PPC_E500
+struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT] = {
+	[MMU_PAGE_4K] = {
+		.shift	= 12,
+	},
+	[MMU_PAGE_2M] = {
+		.shift	= 21,
+	},
+	[MMU_PAGE_4M] = {
+		.shift	= 22,
+	},
+	[MMU_PAGE_16M] = {
+		.shift	= 24,
+	},
+	[MMU_PAGE_64M] = {
+		.shift	= 26,
+	},
+	[MMU_PAGE_256M] = {
+		.shift	= 28,
+	},
+	[MMU_PAGE_1G] = {
+		.shift	= 30,
+	},
+};
+
+static inline int mmu_get_tsize(int psize)
+{
+	return mmu_psize_defs[psize].shift - 10;
+}
+#else
+static inline int mmu_get_tsize(int psize)
+{
+	/* This isn't used on !Book3E for now */
+	return 0;
+}
+#endif
+
+#ifdef CONFIG_PPC_8xx
+struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT] = {
+	[MMU_PAGE_4K] = {
+		.shift	= 12,
+	},
+	[MMU_PAGE_16K] = {
+		.shift	= 14,
+	},
+	[MMU_PAGE_512K] = {
+		.shift	= 19,
+	},
+	[MMU_PAGE_8M] = {
+		.shift	= 23,
+	},
+};
+#endif
+
+#ifdef CONFIG_PPC_E500
+/* next_tlbcam_idx is used to round-robin tlbcam entry assignment */
+DEFINE_PER_CPU(int, next_tlbcam_idx);
+EXPORT_PER_CPU_SYMBOL(next_tlbcam_idx);
+#endif
+
+/*
+ * Base TLB flushing operations:
+ *
+ *  - flush_tlb_mm(mm) flushes the specified mm context TLB's
+ *  - flush_tlb_page(vma, vmaddr) flushes one page
+ *  - flush_tlb_range(vma, start, end) flushes a range of pages
+ *  - flush_tlb_kernel_range(start, end) flushes kernel pages
+ *
+ *  - local_* variants of page and mm only apply to the current
+ *    processor
+ */
+
+#ifndef CONFIG_PPC_8xx
+/*
+ * These are the base non-SMP variants of page and mm flushing
+ */
+void local_flush_tlb_mm(struct mm_struct *mm)
+{
+	unsigned int pid;
+
+	preempt_disable();
+	pid = mm->context.id;
+	if (pid != MMU_NO_CONTEXT)
+		_tlbil_pid(pid);
+	preempt_enable();
+}
+EXPORT_SYMBOL(local_flush_tlb_mm);
+
+void __local_flush_tlb_page(struct mm_struct *mm, unsigned long vmaddr,
+			    int tsize, int ind)
+{
+	unsigned int pid;
+
+	preempt_disable();
+	pid = mm ? mm->context.id : 0;
+	if (pid != MMU_NO_CONTEXT)
+		_tlbil_va(vmaddr, pid, tsize, ind);
+	preempt_enable();
+}
+
+void local_flush_tlb_page(struct vm_area_struct *vma, unsigned long vmaddr)
+{
+	__local_flush_tlb_page(vma ? vma->vm_mm : NULL, vmaddr,
+			       mmu_get_tsize(mmu_virtual_psize), 0);
+}
+EXPORT_SYMBOL(local_flush_tlb_page);
+
+void local_flush_tlb_page_psize(struct mm_struct *mm,
+				unsigned long vmaddr, int psize)
+{
+	__local_flush_tlb_page(mm, vmaddr, mmu_get_tsize(psize), 0);
+}
+EXPORT_SYMBOL(local_flush_tlb_page_psize);
+
+#endif
+
+/*
+ * And here are the SMP non-local implementations
+ */
+#ifdef CONFIG_SMP
+
+static DEFINE_RAW_SPINLOCK(tlbivax_lock);
+
+struct tlb_flush_param {
+	unsigned long addr;
+	unsigned int pid;
+	unsigned int tsize;
+	unsigned int ind;
+};
+
+static void do_flush_tlb_mm_ipi(void *param)
+{
+	struct tlb_flush_param *p = param;
+
+	_tlbil_pid(p ? p->pid : 0);
+}
+
+static void do_flush_tlb_page_ipi(void *param)
+{
+	struct tlb_flush_param *p = param;
+
+	_tlbil_va(p->addr, p->pid, p->tsize, p->ind);
+}
+
+
+/* Note on invalidations and PID:
+ *
+ * We snapshot the PID with preempt disabled. At this point, it can still
+ * change either because:
+ * - our context is being stolen (PID -> NO_CONTEXT) on another CPU
+ * - we are invaliating some target that isn't currently running here
+ *   and is concurrently acquiring a new PID on another CPU
+ * - some other CPU is re-acquiring a lost PID for this mm
+ * etc...
+ *
+ * However, this shouldn't be a problem as we only guarantee
+ * invalidation of TLB entries present prior to this call, so we
+ * don't care about the PID changing, and invalidating a stale PID
+ * is generally harmless.
+ */
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	unsigned int pid;
+
+	preempt_disable();
+	pid = mm->context.id;
+	if (unlikely(pid == MMU_NO_CONTEXT))
+		goto no_context;
+	if (!mm_is_core_local(mm)) {
+		struct tlb_flush_param p = { .pid = pid };
+		/* Ignores smp_processor_id() even if set. */
+		smp_call_function_many(mm_cpumask(mm),
+				       do_flush_tlb_mm_ipi, &p, 1);
+	}
+	_tlbil_pid(pid);
+ no_context:
+	preempt_enable();
+}
+EXPORT_SYMBOL(flush_tlb_mm);
+
+void __flush_tlb_page(struct mm_struct *mm, unsigned long vmaddr,
+		      int tsize, int ind)
+{
+	struct cpumask *cpu_mask;
+	unsigned int pid;
+
+	/*
+	 * This function as well as __local_flush_tlb_page() must only be called
+	 * for user contexts.
+	 */
+	if (WARN_ON(!mm))
+		return;
+
+	preempt_disable();
+	pid = mm->context.id;
+	if (unlikely(pid == MMU_NO_CONTEXT))
+		goto bail;
+	cpu_mask = mm_cpumask(mm);
+	if (!mm_is_core_local(mm)) {
+		/* If broadcast tlbivax is supported, use it */
+		if (mmu_has_feature(MMU_FTR_USE_TLBIVAX_BCAST)) {
+			int lock = mmu_has_feature(MMU_FTR_LOCK_BCAST_INVAL);
+			if (lock)
+				raw_spin_lock(&tlbivax_lock);
+			_tlbivax_bcast(vmaddr, pid, tsize, ind);
+			if (lock)
+				raw_spin_unlock(&tlbivax_lock);
+			goto bail;
+		} else {
+			struct tlb_flush_param p = {
+				.pid = pid,
+				.addr = vmaddr,
+				.tsize = tsize,
+				.ind = ind,
+			};
+			/* Ignores smp_processor_id() even if set in cpu_mask */
+			smp_call_function_many(cpu_mask,
+					       do_flush_tlb_page_ipi, &p, 1);
+		}
+	}
+	_tlbil_va(vmaddr, pid, tsize, ind);
+ bail:
+	preempt_enable();
+}
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long vmaddr)
+{
+#ifdef CONFIG_HUGETLB_PAGE
+	if (vma && is_vm_hugetlb_page(vma))
+		flush_hugetlb_page(vma, vmaddr);
+#endif
+
+	__flush_tlb_page(vma ? vma->vm_mm : NULL, vmaddr,
+			 mmu_get_tsize(mmu_virtual_psize), 0);
+}
+EXPORT_SYMBOL(flush_tlb_page);
+
+#endif /* CONFIG_SMP */
+
+/*
+ * Flush kernel TLB entries in the given range
+ */
+#ifndef CONFIG_PPC_8xx
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+#ifdef CONFIG_SMP
+	preempt_disable();
+	smp_call_function(do_flush_tlb_mm_ipi, NULL, 1);
+	_tlbil_pid(0);
+	preempt_enable();
+#else
+	_tlbil_pid(0);
+#endif
+}
+EXPORT_SYMBOL(flush_tlb_kernel_range);
+#endif
+
+/*
+ * Currently, for range flushing, we just do a full mm flush. This should
+ * be optimized based on a threshold on the size of the range, since
+ * some implementation can stack multiple tlbivax before a tlbsync but
+ * for now, we keep it that way
+ */
+void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+		     unsigned long end)
+
+{
+	if (end - start == PAGE_SIZE && !(start & ~PAGE_MASK))
+		flush_tlb_page(vma, start);
+	else
+		flush_tlb_mm(vma->vm_mm);
+}
+EXPORT_SYMBOL(flush_tlb_range);
+
+void tlb_flush(struct mmu_gather *tlb)
+{
+	flush_tlb_mm(tlb->mm);
+}
+
+#ifndef CONFIG_PPC64
+void __init early_init_mmu(void)
+{
+	unsigned long root = of_get_flat_dt_root();
+
+	if (IS_ENABLED(CONFIG_PPC_47x) && IS_ENABLED(CONFIG_SMP) &&
+	    of_get_flat_dt_prop(root, "cooperative-partition", NULL))
+		mmu_clear_feature(MMU_FTR_USE_TLBIVAX_BCAST);
+}
+#endif /* CONFIG_PPC64 */

linux-7.2.3/arch/powerpc/mm/nohash/tlb_64e.c

@@ -0,0 +1,314 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright 2008,2009 Ben Herrenschmidt <benh@kernel.crashing.org>
+ *                     IBM Corp.
+ *
+ *  Derived from arch/ppc/mm/init.c:
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ *  Modifications by Paul Mackerras (PowerMac) (paulus@cs.anu.edu.au)
+ *  and Cort Dougan (PReP) (cort@cs.nmt.edu)
+ *    Copyright (C) 1996 Paul Mackerras
+ *
+ *  Derived from "arch/i386/mm/init.c"
+ *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
+ */
+
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/pagemap.h>
+#include <linux/memblock.h>
+
+#include <asm/pgalloc.h>
+#include <asm/tlbflush.h>
+#include <asm/tlb.h>
+#include <asm/text-patching.h>
+#include <asm/cputhreads.h>
+
+#include <mm/mmu_decl.h>
+
+/* The variables below are currently only used on 64-bit Book3E
+ * though this will probably be made common with other nohash
+ * implementations at some point
+ */
+static int mmu_pte_psize;	/* Page size used for PTE pages */
+int mmu_vmemmap_psize;		/* Page size used for the virtual mem map */
+int book3e_htw_mode;		/* HW tablewalk?  Value is PPC_HTW_* */
+unsigned long linear_map_top;	/* Top of linear mapping */
+
+
+/*
+ * Number of bytes to add to SPRN_SPRG_TLB_EXFRAME on crit/mcheck/debug
+ * exceptions.  This is used for bolted and e6500 TLB miss handlers which
+ * do not modify this SPRG in the TLB miss code; for other TLB miss handlers,
+ * this is set to zero.
+ */
+int extlb_level_exc;
+
+/*
+ * Handling of virtual linear page tables or indirect TLB entries
+ * flushing when PTE pages are freed
+ */
+void tlb_flush_pgtable(struct mmu_gather *tlb, unsigned long address)
+{
+	int tsize = mmu_psize_defs[mmu_pte_psize].shift - 10;
+
+	if (book3e_htw_mode != PPC_HTW_NONE) {
+		unsigned long start = address & PMD_MASK;
+		unsigned long end = address + PMD_SIZE;
+		unsigned long size = 1UL << mmu_psize_defs[mmu_pte_psize].shift;
+
+		/* This isn't the most optimal, ideally we would factor out the
+		 * while preempt & CPU mask mucking around, or even the IPI but
+		 * it will do for now
+		 */
+		while (start < end) {
+			__flush_tlb_page(tlb->mm, start, tsize, 1);
+			start += size;
+		}
+	} else {
+		unsigned long rmask = 0xf000000000000000ul;
+		unsigned long rid = (address & rmask) | 0x1000000000000000ul;
+		unsigned long vpte = address & ~rmask;
+
+		vpte = (vpte >> (PAGE_SHIFT - 3)) & ~0xffful;
+		vpte |= rid;
+		__flush_tlb_page(tlb->mm, vpte, tsize, 0);
+	}
+}
+
+static void __init setup_page_sizes(void)
+{
+	unsigned int tlb0cfg;
+	unsigned int eptcfg;
+	int psize;
+
+	unsigned int mmucfg = mfspr(SPRN_MMUCFG);
+
+	if ((mmucfg & MMUCFG_MAVN) == MMUCFG_MAVN_V1) {
+		unsigned int tlb1cfg = mfspr(SPRN_TLB1CFG);
+		unsigned int min_pg, max_pg;
+
+		min_pg = (tlb1cfg & TLBnCFG_MINSIZE) >> TLBnCFG_MINSIZE_SHIFT;
+		max_pg = (tlb1cfg & TLBnCFG_MAXSIZE) >> TLBnCFG_MAXSIZE_SHIFT;
+
+		for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) {
+			struct mmu_psize_def *def;
+			unsigned int shift;
+
+			def = &mmu_psize_defs[psize];
+			shift = def->shift;
+
+			if (shift == 0 || shift & 1)
+				continue;
+
+			/* adjust to be in terms of 4^shift Kb */
+			shift = (shift - 10) >> 1;
+
+			if ((shift >= min_pg) && (shift <= max_pg))
+				def->flags |= MMU_PAGE_SIZE_DIRECT;
+		}
+
+		goto out;
+	}
+
+	if ((mmucfg & MMUCFG_MAVN) == MMUCFG_MAVN_V2) {
+		u32 tlb1cfg, tlb1ps;
+
+		tlb0cfg = mfspr(SPRN_TLB0CFG);
+		tlb1cfg = mfspr(SPRN_TLB1CFG);
+		tlb1ps = mfspr(SPRN_TLB1PS);
+		eptcfg = mfspr(SPRN_EPTCFG);
+
+		if ((tlb1cfg & TLBnCFG_IND) && (tlb0cfg & TLBnCFG_PT))
+			book3e_htw_mode = PPC_HTW_E6500;
+
+		/*
+		 * We expect 4K subpage size and unrestricted indirect size.
+		 * The lack of a restriction on indirect size is a Freescale
+		 * extension, indicated by PSn = 0 but SPSn != 0.
+		 */
+		if (eptcfg != 2)
+			book3e_htw_mode = PPC_HTW_NONE;
+
+		for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) {
+			struct mmu_psize_def *def = &mmu_psize_defs[psize];
+
+			if (!def->shift)
+				continue;
+
+			if (tlb1ps & (1U << (def->shift - 10))) {
+				def->flags |= MMU_PAGE_SIZE_DIRECT;
+
+				if (book3e_htw_mode && psize == MMU_PAGE_2M)
+					def->flags |= MMU_PAGE_SIZE_INDIRECT;
+			}
+		}
+
+		goto out;
+	}
+out:
+	/* Cleanup array and print summary */
+	pr_info("MMU: Supported page sizes\n");
+	for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) {
+		struct mmu_psize_def *def = &mmu_psize_defs[psize];
+		const char *__page_type_names[] = {
+			"unsupported",
+			"direct",
+			"indirect",
+			"direct & indirect"
+		};
+		if (def->flags == 0) {
+			def->shift = 0;
+			continue;
+		}
+		pr_info("  %8ld KB as %s\n", 1ul << (def->shift - 10),
+			__page_type_names[def->flags & 0x3]);
+	}
+}
+
+/*
+ * Early initialization of the MMU TLB code
+ */
+static void early_init_this_mmu(void)
+{
+	unsigned int mas4;
+
+	/* Set MAS4 based on page table setting */
+
+	mas4 = 0x4 << MAS4_WIMGED_SHIFT;
+	switch (book3e_htw_mode) {
+	case PPC_HTW_E6500:
+		mas4 |= MAS4_INDD;
+		mas4 |= BOOK3E_PAGESZ_2M << MAS4_TSIZED_SHIFT;
+		mas4 |= MAS4_TLBSELD(1);
+		mmu_pte_psize = MMU_PAGE_2M;
+		break;
+
+	case PPC_HTW_NONE:
+		mas4 |=	BOOK3E_PAGESZ_4K << MAS4_TSIZED_SHIFT;
+		mmu_pte_psize = mmu_virtual_psize;
+		break;
+	}
+	mtspr(SPRN_MAS4, mas4);
+
+	unsigned int num_cams;
+	bool map = true;
+
+	/* use a quarter of the TLBCAM for bolted linear map */
+	num_cams = (mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY) / 4;
+
+	/*
+	 * Only do the mapping once per core, or else the
+	 * transient mapping would cause problems.
+	 */
+#ifdef CONFIG_SMP
+	if (hweight32(get_tensr()) > 1)
+		map = false;
+#endif
+
+	if (map)
+		linear_map_top = map_mem_in_cams(linear_map_top,
+						 num_cams, false, true);
+
+	/* A sync won't hurt us after mucking around with
+	 * the MMU configuration
+	 */
+	mb();
+}
+
+static void __init early_init_mmu_global(void)
+{
+	/*
+	 * Freescale booke only supports 4K pages in TLB0, so use that.
+	 */
+	mmu_vmemmap_psize = MMU_PAGE_4K;
+
+	/* XXX This code only checks for TLB 0 capabilities and doesn't
+	 *     check what page size combos are supported by the HW. It
+	 *     also doesn't handle the case where a separate array holds
+	 *     the IND entries from the array loaded by the PT.
+	 */
+	/* Look for supported page sizes */
+	setup_page_sizes();
+
+	/*
+	 * If we want to use HW tablewalk, enable it by patching the TLB miss
+	 * handlers to branch to the one dedicated to it.
+	 */
+	extlb_level_exc = EX_TLB_SIZE;
+	switch (book3e_htw_mode) {
+	case PPC_HTW_E6500:
+		patch_exception(0x1c0, exc_data_tlb_miss_e6500_book3e);
+		patch_exception(0x1e0, exc_instruction_tlb_miss_e6500_book3e);
+		break;
+	}
+
+	pr_info("MMU: Book3E HW tablewalk %s\n",
+		book3e_htw_mode != PPC_HTW_NONE ? "enabled" : "not supported");
+
+	/* Set the global containing the top of the linear mapping
+	 * for use by the TLB miss code
+	 */
+	linear_map_top = memblock_end_of_DRAM();
+
+	ioremap_bot = IOREMAP_BASE;
+}
+
+static void __init early_mmu_set_memory_limit(void)
+{
+	/*
+	 * Limit memory so we dont have linear faults.
+	 * Unlike memblock_set_current_limit, which limits
+	 * memory available during early boot, this permanently
+	 * reduces the memory available to Linux.  We need to
+	 * do this because highmem is not supported on 64-bit.
+	 */
+	memblock_enforce_memory_limit(linear_map_top);
+
+	memblock_set_current_limit(linear_map_top);
+}
+
+/* boot cpu only */
+void __init early_init_mmu(void)
+{
+	early_init_mmu_global();
+	early_init_this_mmu();
+	early_mmu_set_memory_limit();
+}
+
+void early_init_mmu_secondary(void)
+{
+	early_init_this_mmu();
+}
+
+void setup_initial_memory_limit(phys_addr_t first_memblock_base,
+				phys_addr_t first_memblock_size)
+{
+	/*
+	 * On FSL Embedded 64-bit, usually all RAM is bolted, but with
+	 * unusual memory sizes it's possible for some RAM to not be mapped
+	 * (such RAM is not used at all by Linux, since we don't support
+	 * highmem on 64-bit).  We limit ppc64_rma_size to what would be
+	 * mappable if this memblock is the only one.  Additional memblocks
+	 * can only increase, not decrease, the amount that ends up getting
+	 * mapped.  We still limit max to 1G even if we'll eventually map
+	 * more.  This is due to what the early init code is set up to do.
+	 *
+	 * We crop it to the size of the first MEMBLOCK to
+	 * avoid going over total available memory just in case...
+	 */
+	unsigned long linear_sz;
+	unsigned int num_cams;
+
+	/* use a quarter of the TLBCAM for bolted linear map */
+	num_cams = (mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY) / 4;
+
+	linear_sz = map_mem_in_cams(first_memblock_size, num_cams, true, true);
+	ppc64_rma_size = min_t(u64, linear_sz, 0x40000000);
+
+	/* Finally limit subsequent allocations */
+	memblock_set_current_limit(first_memblock_base + ppc64_rma_size);
+}

linux-7.2.3/arch/powerpc/mm/nohash/tlb_low.S

@@ -0,0 +1,447 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * This file contains low-level functions for performing various
+ * types of TLB invalidations on various processors with no hash
+ * table.
+ *
+ * This file implements the following functions for all no-hash
+ * processors. Some aren't implemented for some variants. Some
+ * are inline in tlbflush.h
+ *
+ *	- tlbil_va
+ *	- tlbil_pid
+ *	- tlbil_all
+ *	- tlbivax_bcast
+ *
+ * Code mostly moved over from misc_32.S
+ *
+ *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
+ *
+ * Partially rewritten by Cort Dougan (cort@cs.nmt.edu)
+ * Paul Mackerras, Kumar Gala and Benjamin Herrenschmidt.
+ */
+
+#include <asm/reg.h>
+#include <asm/page.h>
+#include <asm/cputable.h>
+#include <asm/mmu.h>
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/processor.h>
+#include <asm/bug.h>
+#include <asm/asm-compat.h>
+#include <asm/feature-fixups.h>
+
+#if defined(CONFIG_PPC_8xx)
+
+/*
+ * Nothing to do for 8xx, everything is inline
+ */
+
+#elif defined(CONFIG_44x) /* Includes 47x */
+
+/*
+ * 440 implementation uses tlbsx/we for tlbil_va and a full sweep
+ * of the TLB for everything else.
+ */
+_GLOBAL(__tlbil_va)
+	mfspr	r5,SPRN_MMUCR
+	mfmsr   r10
+
+	/*
+	 * We write 16 bits of STID since 47x supports that much, we
+	 * will never be passed out of bounds values on 440 (hopefully)
+	 */
+	rlwimi  r5,r4,0,16,31
+
+	/* We have to run the search with interrupts disabled, otherwise
+	 * an interrupt which causes a TLB miss can clobber the MMUCR
+	 * between the mtspr and the tlbsx.
+	 *
+	 * Critical and Machine Check interrupts take care of saving
+	 * and restoring MMUCR, so only normal interrupts have to be
+	 * taken care of.
+	 */
+	wrteei	0
+	mtspr	SPRN_MMUCR,r5
+	tlbsx.	r6,0,r3
+	bne	10f
+	sync
+#ifndef CONFIG_PPC_47x
+	/* On 440 There are only 64 TLB entries, so r3 < 64, which means bit
+	 * 22, is clear.  Since 22 is the V bit in the TLB_PAGEID, loading this
+	 * value will invalidate the TLB entry.
+	 */
+	tlbwe	r6,r6,PPC44x_TLB_PAGEID
+#else
+	oris	r7,r6,0x8000	/* specify way explicitly */
+	clrrwi	r4,r3,12	/* get an EPN for the hashing with V = 0 */
+	ori	r4,r4,PPC47x_TLBE_SIZE
+	tlbwe   r4,r7,0		/* write it */
+#endif /* !CONFIG_PPC_47x */
+	isync
+10:	wrtee	r10
+	blr
+
+_GLOBAL(_tlbil_all)
+_GLOBAL(_tlbil_pid)
+#ifndef CONFIG_PPC_47x
+	li	r3,0
+	sync
+
+	/* Load high watermark */
+	lis	r4,tlb_44x_hwater@ha
+	lwz	r5,tlb_44x_hwater@l(r4)
+
+1:	tlbwe	r3,r3,PPC44x_TLB_PAGEID
+	addi	r3,r3,1
+	cmpw	0,r3,r5
+	ble	1b
+
+	isync
+	blr
+#else
+	/* 476 variant. There's not simple way to do this, hopefully we'll
+	 * try to limit the amount of such full invalidates
+	 */
+	mfmsr	r11		/* Interrupts off */
+	wrteei	0
+	li	r3,-1		/* Current set */
+	lis	r10,tlb_47x_boltmap@h
+	ori	r10,r10,tlb_47x_boltmap@l
+	lis	r7,0x8000	/* Specify way explicitly */
+
+	b	9f		/* For each set */
+
+1:	li	r9,4		/* Number of ways */
+	li	r4,0		/* Current way */
+	li	r6,0		/* Default entry value 0 */
+	andi.	r0,r8,1		/* Check if way 0 is bolted */
+	mtctr	r9		/* Load way counter */
+	bne-	3f		/* Bolted, skip loading it */
+
+2:	/* For each way */
+	or	r5,r3,r4	/* Make way|index for tlbre */
+	rlwimi	r5,r5,16,8,15	/* Copy index into position */
+	tlbre	r6,r5,0		/* Read entry */
+3:	addis	r4,r4,0x2000	/* Next way */
+	andi.	r0,r6,PPC47x_TLB0_VALID /* Valid entry ? */
+	beq	4f		/* Nope, skip it */
+	rlwimi	r7,r5,0,1,2	/* Insert way number */
+	rlwinm	r6,r6,0,21,19	/* Clear V */
+	tlbwe   r6,r7,0		/* Write it */
+4:	bdnz	2b		/* Loop for each way */
+	srwi	r8,r8,1		/* Next boltmap bit */
+9:	cmpwi	cr1,r3,255	/* Last set done ? */
+	addi	r3,r3,1		/* Next set */
+	beq	cr1,1f		/* End of loop */
+	andi.	r0,r3,0x1f	/* Need to load a new boltmap word ? */
+	bne	1b		/* No, loop */
+	lwz	r8,0(r10)	/* Load boltmap entry */
+	addi	r10,r10,4	/* Next word */
+	b	1b		/* Then loop */
+1:	isync			/* Sync shadows */
+	wrtee	r11
+	blr
+#endif /* !CONFIG_PPC_47x */
+
+#ifdef CONFIG_PPC_47x
+
+/*
+ * _tlbivax_bcast is only on 47x. We don't bother doing a runtime
+ * check though, it will blow up soon enough if we mistakenly try
+ * to use it on a 440.
+ */
+_GLOBAL(_tlbivax_bcast)
+	mfspr	r5,SPRN_MMUCR
+	mfmsr	r10
+	rlwimi	r5,r4,0,16,31
+	wrteei	0
+	mtspr	SPRN_MMUCR,r5
+	isync
+	PPC_TLBIVAX(0, R3)
+	isync
+	mbar
+	tlbsync
+BEGIN_FTR_SECTION
+	b	1f
+END_FTR_SECTION_IFSET(CPU_FTR_476_DD2)
+	sync
+	wrtee	r10
+	blr
+/*
+ * DD2 HW could hang if in instruction fetch happens before msync completes.
+ * Touch enough instruction cache lines to ensure cache hits
+ */
+1:	mflr	r9
+	bcl	20,31,$+4
+2:	mflr	r6
+	li	r7,32
+	PPC_ICBT(0,R6,R7)		/* touch next cache line */
+	add	r6,r6,r7
+	PPC_ICBT(0,R6,R7)		/* touch next cache line */
+	add	r6,r6,r7
+	PPC_ICBT(0,R6,R7)		/* touch next cache line */
+	sync
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	mtlr	r9
+	wrtee	r10
+	blr
+#endif /* CONFIG_PPC_47x */
+
+#elif defined(CONFIG_PPC_85xx)
+/*
+ * FSL BookE implementations.
+ *
+ * Since feature sections are using _SECTION_ELSE we need
+ * to have the larger code path before the _SECTION_ELSE
+ */
+
+/*
+ * Flush MMU TLB on the local processor
+ */
+_GLOBAL(_tlbil_all)
+BEGIN_MMU_FTR_SECTION
+	li	r3,(MMUCSR0_TLBFI)@l
+	mtspr	SPRN_MMUCSR0, r3
+1:
+	mfspr	r3,SPRN_MMUCSR0
+	andi.	r3,r3,MMUCSR0_TLBFI@l
+	bne	1b
+MMU_FTR_SECTION_ELSE
+	PPC_TLBILX_ALL(0,R0)
+ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_USE_TLBILX)
+	msync
+	isync
+	blr
+
+_GLOBAL(_tlbil_pid)
+BEGIN_MMU_FTR_SECTION
+	slwi	r3,r3,16
+	mfmsr	r10
+	wrteei	0
+	mfspr	r4,SPRN_MAS6	/* save MAS6 */
+	mtspr	SPRN_MAS6,r3
+	PPC_TLBILX_PID(0,R0)
+	mtspr	SPRN_MAS6,r4	/* restore MAS6 */
+	wrtee	r10
+MMU_FTR_SECTION_ELSE
+	li	r3,(MMUCSR0_TLBFI)@l
+	mtspr	SPRN_MMUCSR0, r3
+1:
+	mfspr	r3,SPRN_MMUCSR0
+	andi.	r3,r3,MMUCSR0_TLBFI@l
+	bne	1b
+ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_USE_TLBILX)
+	msync
+	isync
+	blr
+
+/*
+ * Flush MMU TLB for a particular address, but only on the local processor
+ * (no broadcast)
+ */
+_GLOBAL(__tlbil_va)
+	mfmsr	r10
+	wrteei	0
+	slwi	r4,r4,16
+	ori	r4,r4,(MAS6_ISIZE(BOOK3E_PAGESZ_4K))@l
+	mtspr	SPRN_MAS6,r4		/* assume AS=0 for now */
+BEGIN_MMU_FTR_SECTION
+	tlbsx	0,r3
+	mfspr	r4,SPRN_MAS1		/* check valid */
+	andis.	r3,r4,MAS1_VALID@h
+	beq	1f
+	rlwinm	r4,r4,0,1,31
+	mtspr	SPRN_MAS1,r4
+	tlbwe
+MMU_FTR_SECTION_ELSE
+	PPC_TLBILX_VA(0,R3)
+ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_USE_TLBILX)
+	msync
+	isync
+1:	wrtee	r10
+	blr
+#elif defined(CONFIG_PPC_BOOK3E_64)
+/*
+ * New Book3E (>= 2.06) implementation
+ *
+ * Note: We may be able to get away without the interrupt masking stuff
+ * if we save/restore MAS6 on exceptions that might modify it
+ */
+_GLOBAL(_tlbil_pid)
+	slwi	r4,r3,MAS6_SPID_SHIFT
+	mfmsr	r10
+	wrteei	0
+	mtspr	SPRN_MAS6,r4
+	PPC_TLBILX_PID(0,R0)
+	wrtee	r10
+	msync
+	isync
+	blr
+
+_GLOBAL(_tlbil_pid_noind)
+	slwi	r4,r3,MAS6_SPID_SHIFT
+	mfmsr	r10
+	ori	r4,r4,MAS6_SIND
+	wrteei	0
+	mtspr	SPRN_MAS6,r4
+	PPC_TLBILX_PID(0,R0)
+	wrtee	r10
+	msync
+	isync
+	blr
+
+_GLOBAL(_tlbil_all)
+	PPC_TLBILX_ALL(0,R0)
+	msync
+	isync
+	blr
+
+_GLOBAL(_tlbil_va)
+	mfmsr	r10
+	wrteei	0
+	cmpwi	cr0,r6,0
+	slwi	r4,r4,MAS6_SPID_SHIFT
+	rlwimi	r4,r5,MAS6_ISIZE_SHIFT,MAS6_ISIZE_MASK
+	beq	1f
+	rlwimi	r4,r6,MAS6_SIND_SHIFT,MAS6_SIND
+1:	mtspr	SPRN_MAS6,r4		/* assume AS=0 for now */
+	PPC_TLBILX_VA(0,R3)
+	msync
+	isync
+	wrtee	r10
+	blr
+
+_GLOBAL(_tlbivax_bcast)
+	mfmsr	r10
+	wrteei	0
+	cmpwi	cr0,r6,0
+	slwi	r4,r4,MAS6_SPID_SHIFT
+	rlwimi	r4,r5,MAS6_ISIZE_SHIFT,MAS6_ISIZE_MASK
+	beq	1f
+	rlwimi	r4,r6,MAS6_SIND_SHIFT,MAS6_SIND
+1:	mtspr	SPRN_MAS6,r4		/* assume AS=0 for now */
+	PPC_TLBIVAX(0,R3)
+	mbar
+	tlbsync
+	sync
+	wrtee	r10
+	blr
+#else
+#error Unsupported processor type !
+#endif
+
+#if defined(CONFIG_PPC_E500)
+/*
+ * extern void loadcam_entry(unsigned int index)
+ *
+ * Load TLBCAM[index] entry in to the L2 CAM MMU
+ * Must preserve r7, r8, r9, r10, r11, r12
+ */
+_GLOBAL(loadcam_entry)
+	mflr	r5
+	LOAD_REG_ADDR_PIC(r4, TLBCAM)
+	mtlr	r5
+	mulli	r5,r3,TLBCAM_SIZE
+	add	r3,r5,r4
+	lwz	r4,TLBCAM_MAS0(r3)
+	mtspr	SPRN_MAS0,r4
+	lwz	r4,TLBCAM_MAS1(r3)
+	mtspr	SPRN_MAS1,r4
+	PPC_LL	r4,TLBCAM_MAS2(r3)
+	mtspr	SPRN_MAS2,r4
+	lwz	r4,TLBCAM_MAS3(r3)
+	mtspr	SPRN_MAS3,r4
+BEGIN_MMU_FTR_SECTION
+	lwz	r4,TLBCAM_MAS7(r3)
+	mtspr	SPRN_MAS7,r4
+END_MMU_FTR_SECTION_IFSET(MMU_FTR_BIG_PHYS)
+	isync
+	tlbwe
+	isync
+	blr
+
+/*
+ * Load multiple TLB entries at once, using an alternate-space
+ * trampoline so that we don't have to care about whether the same
+ * TLB entry maps us before and after.
+ *
+ * r3 = first entry to write
+ * r4 = number of entries to write
+ * r5 = temporary tlb entry (0 means no switch to AS1)
+ */
+_GLOBAL(loadcam_multi)
+	mflr	r8
+	/* Don't switch to AS=1 if already there */
+	mfmsr	r11
+	andi.	r11,r11,MSR_IS
+	bne	10f
+	mr.	r12, r5
+	beq	10f
+
+	/*
+	 * Set up temporary TLB entry that is the same as what we're
+	 * running from, but in AS=1.
+	 */
+	bcl	20,31,$+4
+1:	mflr	r6
+	tlbsx	0,r8
+	mfspr	r6,SPRN_MAS1
+	ori	r6,r6,MAS1_TS
+	mtspr	SPRN_MAS1,r6
+	mfspr	r6,SPRN_MAS0
+	rlwimi	r6,r5,MAS0_ESEL_SHIFT,MAS0_ESEL_MASK
+	mr	r7,r5
+	mtspr	SPRN_MAS0,r6
+	isync
+	tlbwe
+	isync
+
+	/* Switch to AS=1 */
+	mfmsr	r6
+	ori	r6,r6,MSR_IS|MSR_DS
+	mtmsr	r6
+	isync
+
+10:
+	mr	r9,r3
+	add	r10,r3,r4
+2:	bl	loadcam_entry
+	addi	r9,r9,1
+	cmpw	r9,r10
+	mr	r3,r9
+	blt	2b
+
+	/* Don't return to AS=0 if we were in AS=1 at function start */
+	andi.	r11,r11,MSR_IS
+	bne	3f
+	cmpwi	r12, 0
+	beq	3f
+
+	/* Return to AS=0 and clear the temporary entry */
+	mfmsr	r6
+	rlwinm.	r6,r6,0,~(MSR_IS|MSR_DS)
+	mtmsr	r6
+	isync
+
+	li	r6,0
+	mtspr	SPRN_MAS1,r6
+	rlwinm	r6,r7,MAS0_ESEL_SHIFT,MAS0_ESEL_MASK
+	oris	r6,r6,MAS0_TLBSEL(1)@h
+	mtspr	SPRN_MAS0,r6
+	isync
+	tlbwe
+	isync
+
+3:
+	mtlr	r8
+	blr
+#endif

linux-7.2.3/arch/powerpc/mm/nohash/tlb_low_64e.S

@@ -0,0 +1,743 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ *  Low level TLB miss handlers for Book3E
+ *
+ *  Copyright (C) 2008-2009
+ *      Ben. Herrenschmidt (benh@kernel.crashing.org), IBM Corp.
+ */
+
+#include <linux/pgtable.h>
+#include <asm/processor.h>
+#include <asm/reg.h>
+#include <asm/page.h>
+#include <asm/mmu.h>
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/cputable.h>
+#include <asm/exception-64e.h>
+#include <asm/ppc-opcode.h>
+#include <asm/kvm_asm.h>
+#include <asm/kvm_booke_hv_asm.h>
+#include <asm/feature-fixups.h>
+
+#define VPTE_PMD_SHIFT	(PTE_INDEX_SIZE)
+#define VPTE_PUD_SHIFT	(VPTE_PMD_SHIFT + PMD_INDEX_SIZE)
+#define VPTE_PGD_SHIFT	(VPTE_PUD_SHIFT + PUD_INDEX_SIZE)
+#define VPTE_INDEX_SIZE (VPTE_PGD_SHIFT + PGD_INDEX_SIZE)
+
+/**********************************************************************
+ *                                                                    *
+ * TLB miss handling for Book3E with a bolted linear mapping          *
+ * No virtual page table, no nested TLB misses                        *
+ *                                                                    *
+ **********************************************************************/
+
+/*
+ * Note that, unlike non-bolted handlers, TLB_EXFRAME is not
+ * modified by the TLB miss handlers themselves, since the TLB miss
+ * handler code will not itself cause a recursive TLB miss.
+ *
+ * TLB_EXFRAME will be modified when crit/mc/debug exceptions are
+ * entered/exited.
+ */
+.macro tlb_prolog_bolted intnum addr
+	mtspr	SPRN_SPRG_GEN_SCRATCH,r12
+	mfspr	r12,SPRN_SPRG_TLB_EXFRAME
+	std	r13,EX_TLB_R13(r12)
+	std	r10,EX_TLB_R10(r12)
+	mfspr	r13,SPRN_SPRG_PACA
+
+	mfcr	r10
+	std	r11,EX_TLB_R11(r12)
+#ifdef CONFIG_KVM_BOOKE_HV
+BEGIN_FTR_SECTION
+	mfspr	r11, SPRN_SRR1
+END_FTR_SECTION_IFSET(CPU_FTR_EMB_HV)
+#endif
+	DO_KVM	\intnum, SPRN_SRR1
+	std	r16,EX_TLB_R16(r12)
+	mfspr	r16,\addr		/* get faulting address */
+	std	r14,EX_TLB_R14(r12)
+	ld	r14,PACAPGD(r13)
+	std	r15,EX_TLB_R15(r12)
+	std	r10,EX_TLB_CR(r12)
+START_BTB_FLUSH_SECTION
+	mfspr r11, SPRN_SRR1
+	andi. r10,r11,MSR_PR
+	beq 1f
+	BTB_FLUSH(r10)
+1:
+END_BTB_FLUSH_SECTION
+	std	r7,EX_TLB_R7(r12)
+.endm
+
+.macro tlb_epilog_bolted
+	ld	r14,EX_TLB_CR(r12)
+	ld	r7,EX_TLB_R7(r12)
+	ld	r10,EX_TLB_R10(r12)
+	ld	r11,EX_TLB_R11(r12)
+	ld	r13,EX_TLB_R13(r12)
+	mtcr	r14
+	ld	r14,EX_TLB_R14(r12)
+	ld	r15,EX_TLB_R15(r12)
+	ld	r16,EX_TLB_R16(r12)
+	mfspr	r12,SPRN_SPRG_GEN_SCRATCH
+.endm
+
+/* Data TLB miss */
+	START_EXCEPTION(data_tlb_miss_bolted)
+	tlb_prolog_bolted BOOKE_INTERRUPT_DTLB_MISS SPRN_DEAR
+
+	/* We need _PAGE_PRESENT and  _PAGE_ACCESSED set */
+
+	/* We do the user/kernel test for the PID here along with the RW test
+	 */
+	/* We pre-test some combination of permissions to avoid double
+	 * faults:
+	 *
+	 * We move the ESR:ST bit into the position of _PAGE_BAP_SW in the PTE
+	 * ESR_ST   is 0x00800000
+	 * _PAGE_BAP_SW is 0x00000010
+	 * So the shift is >> 19. This tests for supervisor writeability.
+	 * If the page happens to be supervisor writeable and not user
+	 * writeable, we will take a new fault later, but that should be
+	 * a rare enough case.
+	 *
+	 * We also move ESR_ST in _PAGE_DIRTY position
+	 * _PAGE_DIRTY is 0x00001000 so the shift is >> 11
+	 *
+	 * MAS1 is preset for all we need except for TID that needs to
+	 * be cleared for kernel translations
+	 */
+
+	mfspr	r11,SPRN_ESR
+
+	srdi	r15,r16,60		/* get region */
+	rldicl.	r10,r16,64-PGTABLE_EADDR_SIZE,PGTABLE_EADDR_SIZE+4
+	bne-	dtlb_miss_fault_bolted	/* Bail if fault addr is invalid */
+
+	rlwinm	r10,r11,32-19,27,27
+	rlwimi	r10,r11,32-16,19,19
+	cmpwi	r15,0			/* user vs kernel check */
+	ori	r10,r10,_PAGE_PRESENT
+	oris	r11,r10,_PAGE_ACCESSED@h
+
+	bne	tlb_miss_kernel_bolted
+
+tlb_miss_user_bolted:
+#ifdef CONFIG_PPC_KUAP
+	mfspr	r10,SPRN_MAS1
+	rlwinm.	r10,r10,0,0x3fff0000
+	beq-	tlb_miss_fault_bolted /* KUAP fault */
+#endif
+
+tlb_miss_common_bolted:
+/*
+ * This is the guts of the TLB miss handler for bolted-linear.
+ * We are entered with:
+ *
+ * r16 = faulting address
+ * r15 = crap (free to use)
+ * r14 = page table base
+ * r13 = PACA
+ * r11 = PTE permission mask
+ * r10 = crap (free to use)
+ */
+	rldicl	r15,r16,64-PGDIR_SHIFT+3,64-PGD_INDEX_SIZE-3
+	cmpldi	cr0,r14,0
+	clrrdi	r15,r15,3
+	beq	tlb_miss_fault_bolted	/* No PGDIR, bail */
+
+	ldx	r14,r14,r15		/* grab pgd entry */
+
+	rldicl	r15,r16,64-PUD_SHIFT+3,64-PUD_INDEX_SIZE-3
+	clrrdi	r15,r15,3
+	cmpdi	cr0,r14,0
+	bge	tlb_miss_fault_bolted	/* Bad pgd entry or hugepage; bail */
+	ldx	r14,r14,r15		/* grab pud entry */
+
+	rldicl	r15,r16,64-PMD_SHIFT+3,64-PMD_INDEX_SIZE-3
+	clrrdi	r15,r15,3
+	cmpdi	cr0,r14,0
+	bge	tlb_miss_fault_bolted
+	ldx	r14,r14,r15		/* Grab pmd entry */
+
+	rldicl	r15,r16,64-PAGE_SHIFT+3,64-PTE_INDEX_SIZE-3
+	clrrdi	r15,r15,3
+	cmpdi	cr0,r14,0
+	bge	tlb_miss_fault_bolted
+	ldx	r14,r14,r15		/* Grab PTE, normal (!huge) page */
+
+	/* Check if required permissions are met */
+	andc.	r15,r11,r14
+	rldicr	r15,r14,64-(PTE_RPN_SHIFT-PAGE_SHIFT),63-PAGE_SHIFT
+	bne-	tlb_miss_fault_bolted
+
+	/* Now we build the MAS:
+	 *
+	 * MAS 0   :	Fully setup with defaults in MAS4 and TLBnCFG
+	 * MAS 1   :	Almost fully setup
+	 *               - PID already updated by caller if necessary
+	 *               - TSIZE need change if !base page size, not
+	 *                 yet implemented for now
+	 * MAS 2   :	Defaults not useful, need to be redone
+	 * MAS 3+7 :	Needs to be done
+	 */
+	clrrdi	r11,r16,12		/* Clear low crap in EA */
+	clrldi	r15,r15,12		/* Clear crap at the top */
+	rlwimi	r11,r14,32-19,27,31	/* Insert WIMGE */
+	rlwimi	r15,r14,32-8,22,25	/* Move in U bits */
+	mtspr	SPRN_MAS2,r11
+	andi.	r11,r14,_PAGE_DIRTY
+	rlwimi	r15,r14,32-2,26,31	/* Move in BAP bits */
+
+	/* Mask out SW and UW if !DIRTY (XXX optimize this !) */
+	bne	1f
+	li	r11,MAS3_SW|MAS3_UW
+	andc	r15,r15,r11
+1:
+	mtspr	SPRN_MAS7_MAS3,r15
+	tlbwe
+
+tlb_miss_done_bolted:
+	tlb_epilog_bolted
+	rfi
+
+itlb_miss_kernel_bolted:
+	li	r11,_PAGE_PRESENT|_PAGE_BAP_SX	/* Base perm */
+	oris	r11,r11,_PAGE_ACCESSED@h
+tlb_miss_kernel_bolted:
+	mfspr	r10,SPRN_MAS1
+	ld	r14,PACA_KERNELPGD(r13)
+	srdi	r15,r16,44		/* get kernel region */
+	andi.	r15,r15,1		/* Check for vmalloc region */
+	rlwinm	r10,r10,0,16,1		/* Clear TID */
+	mtspr	SPRN_MAS1,r10
+	bne+	tlb_miss_common_bolted
+
+tlb_miss_fault_bolted:
+	/* We need to check if it was an instruction miss */
+	andi.	r10,r11,_PAGE_BAP_UX|_PAGE_BAP_SX
+	bne	itlb_miss_fault_bolted
+dtlb_miss_fault_bolted:
+	tlb_epilog_bolted
+	b	exc_data_storage_book3e
+itlb_miss_fault_bolted:
+	tlb_epilog_bolted
+	b	exc_instruction_storage_book3e
+
+/* Instruction TLB miss */
+	START_EXCEPTION(instruction_tlb_miss_bolted)
+	tlb_prolog_bolted BOOKE_INTERRUPT_ITLB_MISS SPRN_SRR0
+
+	rldicl.	r10,r16,64-PGTABLE_EADDR_SIZE,PGTABLE_EADDR_SIZE+4
+	srdi	r15,r16,60		/* get region */
+	bne-	itlb_miss_fault_bolted
+
+	li	r11,_PAGE_PRESENT|_PAGE_BAP_UX	/* Base perm */
+
+	/* We do the user/kernel test for the PID here along with the RW test
+	 */
+
+	cmpldi	cr0,r15,0			/* Check for user region */
+	oris	r11,r11,_PAGE_ACCESSED@h
+	beq	tlb_miss_user_bolted
+	b	itlb_miss_kernel_bolted
+
+/*
+ * TLB miss handling for e6500 and derivatives, using hardware tablewalk.
+ *
+ * Linear mapping is bolted: no virtual page table or nested TLB misses
+ * Indirect entries in TLB1, hardware loads resulting direct entries
+ *    into TLB0
+ * No HES or NV hint on TLB1, so we need to do software round-robin
+ * No tlbsrx. so we need a spinlock, and we have to deal
+ *    with MAS-damage caused by tlbsx
+ * 4K pages only
+ */
+
+	START_EXCEPTION(instruction_tlb_miss_e6500)
+	tlb_prolog_bolted BOOKE_INTERRUPT_ITLB_MISS SPRN_SRR0
+
+	ld	r11,PACA_TCD_PTR(r13)
+	srdi.	r15,r16,60		/* get region */
+	ori	r16,r16,1
+
+	bne	tlb_miss_kernel_e6500	/* user/kernel test */
+
+	b	tlb_miss_common_e6500
+
+	START_EXCEPTION(data_tlb_miss_e6500)
+	tlb_prolog_bolted BOOKE_INTERRUPT_DTLB_MISS SPRN_DEAR
+
+	ld	r11,PACA_TCD_PTR(r13)
+	srdi.	r15,r16,60		/* get region */
+	rldicr	r16,r16,0,62
+
+	bne	tlb_miss_kernel_e6500	/* user vs kernel check */
+
+/*
+ * This is the guts of the TLB miss handler for e6500 and derivatives.
+ * We are entered with:
+ *
+ * r16 = page of faulting address (low bit 0 if data, 1 if instruction)
+ * r15 = crap (free to use)
+ * r14 = page table base
+ * r13 = PACA
+ * r11 = tlb_per_core ptr
+ * r10 = crap (free to use)
+ * r7  = esel_next
+ */
+tlb_miss_common_e6500:
+	crmove	cr2*4+2,cr0*4+2		/* cr2.eq != 0 if kernel address */
+
+BEGIN_FTR_SECTION		/* CPU_FTR_SMT */
+	/*
+	 * Search if we already have an indirect entry for that virtual
+	 * address, and if we do, bail out.
+	 *
+	 * MAS6:IND should be already set based on MAS4
+	 */
+	lhz	r10,PACAPACAINDEX(r13)
+	addi	r10,r10,1
+	crclr	cr1*4+eq	/* set cr1.eq = 0 for non-recursive */
+1:	lbarx	r15,0,r11
+	cmpdi	r15,0
+	bne	2f
+	stbcx.	r10,0,r11
+	bne	1b
+3:
+	.subsection 1
+2:	cmpd	cr1,r15,r10	/* recursive lock due to mcheck/crit/etc? */
+	beq	cr1,3b		/* unlock will happen if cr1.eq = 0 */
+10:	lbz	r15,0(r11)
+	cmpdi	r15,0
+	bne	10b
+	b	1b
+	.previous
+END_FTR_SECTION_IFSET(CPU_FTR_SMT)
+
+	lbz	r7,TCD_ESEL_NEXT(r11)
+
+BEGIN_FTR_SECTION		/* CPU_FTR_SMT */
+	/*
+	 * Erratum A-008139 says that we can't use tlbwe to change
+	 * an indirect entry in any way (including replacing or
+	 * invalidating) if the other thread could be in the process
+	 * of a lookup.  The workaround is to invalidate the entry
+	 * with tlbilx before overwriting.
+	 */
+
+	rlwinm	r10,r7,16,0xff0000
+	oris	r10,r10,MAS0_TLBSEL(1)@h
+	mtspr	SPRN_MAS0,r10
+	isync
+	tlbre
+	mfspr	r15,SPRN_MAS1
+	andis.	r15,r15,MAS1_VALID@h
+	beq	5f
+
+BEGIN_FTR_SECTION_NESTED(532)
+	mfspr	r10,SPRN_MAS8
+	rlwinm	r10,r10,0,0x80000fff  /* tgs,tlpid -> sgs,slpid */
+	mtspr	SPRN_MAS5,r10
+END_FTR_SECTION_NESTED(CPU_FTR_EMB_HV,CPU_FTR_EMB_HV,532)
+
+	mfspr	r10,SPRN_MAS1
+	rlwinm	r15,r10,0,0x3fff0000  /* tid -> spid */
+	rlwimi	r15,r10,20,0x00000003 /* ind,ts -> sind,sas */
+	mfspr	r10,SPRN_MAS6
+	mtspr	SPRN_MAS6,r15
+
+	mfspr	r15,SPRN_MAS2
+	isync
+	PPC_TLBILX_VA(0,R15)
+	isync
+
+	mtspr	SPRN_MAS6,r10
+
+5:
+BEGIN_FTR_SECTION_NESTED(532)
+	li	r10,0
+	mtspr	SPRN_MAS8,r10
+	mtspr	SPRN_MAS5,r10
+END_FTR_SECTION_NESTED(CPU_FTR_EMB_HV,CPU_FTR_EMB_HV,532)
+
+	tlbsx	0,r16
+	mfspr	r10,SPRN_MAS1
+	andis.	r15,r10,MAS1_VALID@h
+	bne	tlb_miss_done_e6500
+FTR_SECTION_ELSE
+	mfspr	r10,SPRN_MAS1
+ALT_FTR_SECTION_END_IFSET(CPU_FTR_SMT)
+
+	oris	r10,r10,MAS1_VALID@h
+	beq	cr2,4f
+	rlwinm	r10,r10,0,16,1		/* Clear TID */
+4:	mtspr	SPRN_MAS1,r10
+
+	/* Now, we need to walk the page tables. First check if we are in
+	 * range.
+	 */
+	rldicl.	r10,r16,64-PGTABLE_EADDR_SIZE,PGTABLE_EADDR_SIZE+4
+	bne-	tlb_miss_fault_e6500
+
+	rldicl	r15,r16,64-PGDIR_SHIFT+3,64-PGD_INDEX_SIZE-3
+	cmpldi	cr0,r14,0
+	clrrdi	r15,r15,3
+	beq-	tlb_miss_fault_e6500 /* No PGDIR, bail */
+	ldx	r14,r14,r15		/* grab pgd entry */
+
+	rldicl	r15,r16,64-PUD_SHIFT+3,64-PUD_INDEX_SIZE-3
+	clrrdi	r15,r15,3
+	cmpdi	cr0,r14,0
+	bge	tlb_miss_huge_e6500	/* Bad pgd entry or hugepage; bail */
+	ldx	r14,r14,r15		/* grab pud entry */
+
+	rldicl	r15,r16,64-PMD_SHIFT+3,64-PMD_INDEX_SIZE-3
+	clrrdi	r15,r15,3
+	cmpdi	cr0,r14,0
+	bge	tlb_miss_huge_e6500
+	ldx	r14,r14,r15		/* Grab pmd entry */
+
+	mfspr	r10,SPRN_MAS0
+	cmpdi	cr0,r14,0
+	bge	tlb_miss_huge_e6500
+
+	/* Now we build the MAS for a 2M indirect page:
+	 *
+	 * MAS 0   :	ESEL needs to be filled by software round-robin
+	 * MAS 1   :	Fully set up
+	 *               - PID already updated by caller if necessary
+	 *               - TSIZE for now is base ind page size always
+	 *               - TID already cleared if necessary
+	 * MAS 2   :	Default not 2M-aligned, need to be redone
+	 * MAS 3+7 :	Needs to be done
+	 */
+
+	ori	r14,r14,(BOOK3E_PAGESZ_4K << MAS3_SPSIZE_SHIFT)
+	mtspr	SPRN_MAS7_MAS3,r14
+
+	clrrdi	r15,r16,21		/* make EA 2M-aligned */
+	mtspr	SPRN_MAS2,r15
+
+tlb_miss_huge_done_e6500:
+	lbz	r16,TCD_ESEL_MAX(r11)
+	lbz	r14,TCD_ESEL_FIRST(r11)
+	rlwimi	r10,r7,16,0x00ff0000	/* insert esel_next into MAS0 */
+	addi	r7,r7,1			/* increment esel_next */
+	mtspr	SPRN_MAS0,r10
+	cmpw	r7,r16
+	iseleq	r7,r14,r7		/* if next == last use first */
+	stb	r7,TCD_ESEL_NEXT(r11)
+
+	tlbwe
+
+tlb_miss_done_e6500:
+	.macro	tlb_unlock_e6500
+BEGIN_FTR_SECTION
+	beq	cr1,1f		/* no unlock if lock was recursively grabbed */
+	li	r15,0
+	isync
+	stb	r15,0(r11)
+1:
+END_FTR_SECTION_IFSET(CPU_FTR_SMT)
+	.endm
+
+	tlb_unlock_e6500
+	tlb_epilog_bolted
+	rfi
+
+tlb_miss_huge_e6500:
+	beq	tlb_miss_fault_e6500
+	rlwinm	r15,r14,32-_PAGE_PSIZE_SHIFT,0x1e
+
+	/*
+	 * Now we build the MAS for a huge page.
+	 *
+	 * MAS 0   :	ESEL needs to be filled by software round-robin
+	 *		 - can be handled by indirect code
+	 * MAS 1   :	Need to clear IND and set TSIZE
+	 * MAS 2,3+7:	Needs to be redone similar to non-tablewalk handler
+	 */
+
+	mfspr	r10,SPRN_MAS1
+	rlwinm	r10,r10,0,~MAS1_IND
+	rlwimi	r10,r15,MAS1_TSIZE_SHIFT,MAS1_TSIZE_MASK
+	mtspr	SPRN_MAS1,r10
+
+	li	r10,-0x400
+	sld	r15,r10,r15		/* Generate mask based on size */
+	and	r10,r16,r15
+	rldicr	r15,r14,64-(PTE_RPN_SHIFT-PAGE_SHIFT),63-PAGE_SHIFT
+	rlwimi	r10,r14,32-19,27,31	/* Insert WIMGE */
+	clrldi	r15,r15,PAGE_SHIFT	/* Clear crap at the top */
+	rlwimi	r15,r14,32-8,22,25	/* Move in U bits */
+	mtspr	SPRN_MAS2,r10
+	andi.	r10,r14,_PAGE_DIRTY
+	rlwimi	r15,r14,32-2,26,31	/* Move in BAP bits */
+
+	/* Mask out SW and UW if !DIRTY (XXX optimize this !) */
+	bne	1f
+	li	r10,MAS3_SW|MAS3_UW
+	andc	r15,r15,r10
+1:
+	mtspr	SPRN_MAS7_MAS3,r15
+
+	mfspr	r10,SPRN_MAS0
+	b	tlb_miss_huge_done_e6500
+
+tlb_miss_kernel_e6500:
+	ld	r14,PACA_KERNELPGD(r13)
+	srdi	r15,r16,44		/* get kernel region */
+	xoris	r15,r15,0xc		/* Check for vmalloc region */
+	cmplwi	cr1,r15,1
+	beq+	cr1,tlb_miss_common_e6500
+
+tlb_miss_fault_e6500:
+	tlb_unlock_e6500
+	/* We need to check if it was an instruction miss */
+	andi.	r16,r16,1
+	bne	itlb_miss_fault_e6500
+dtlb_miss_fault_e6500:
+	tlb_epilog_bolted
+	b	exc_data_storage_book3e
+itlb_miss_fault_e6500:
+	tlb_epilog_bolted
+	b	exc_instruction_storage_book3e
+
+/*
+ * This is the guts of the second-level TLB miss handler for direct
+ * misses. We are entered with:
+ *
+ * r16 = virtual page table faulting address
+ * r15 = region (top 4 bits of address)
+ * r14 = crap (free to use)
+ * r13 = PACA
+ * r12 = TLB exception frame in PACA
+ * r11 = crap (free to use)
+ * r10 = crap (free to use)
+ *
+ * Note that this should only ever be called as a second level handler
+ * with the current scheme when using SW load.
+ * That means we can always get the original fault DEAR at
+ * EX_TLB_DEAR-EX_TLB_SIZE(r12)
+ *
+ * It can be re-entered by the linear mapping miss handler. However, to
+ * avoid too much complication, it will restart the whole fault at level
+ * 0 so we don't care too much about clobbers
+ *
+ * XXX That code was written back when we couldn't clobber r14. We can now,
+ * so we could probably optimize things a bit
+ */
+virt_page_table_tlb_miss:
+	/* Are we hitting a kernel page table ? */
+	srdi	r15,r16,60
+	andi.	r10,r15,0x8
+
+	/* The cool thing now is that r10 contains 0 for user and 8 for kernel,
+	 * and we happen to have the swapper_pg_dir at offset 8 from the user
+	 * pgdir in the PACA :-).
+	 */
+	add	r11,r10,r13
+
+	/* If kernel, we need to clear MAS1 TID */
+	beq	1f
+	/* XXX replace the RMW cycles with immediate loads + writes */
+	mfspr	r10,SPRN_MAS1
+	rlwinm	r10,r10,0,16,1			/* Clear TID */
+	mtspr	SPRN_MAS1,r10
+#ifdef CONFIG_PPC_KUAP
+	b	2f
+1:
+	mfspr	r10,SPRN_MAS1
+	rlwinm.	r10,r10,0,0x3fff0000
+	beq-	virt_page_table_tlb_miss_fault /* KUAP fault */
+2:
+#else
+1:
+#endif
+
+	/* Now, we need to walk the page tables. First check if we are in
+	 * range.
+	 */
+	rldicl	r10,r16,64-(VPTE_INDEX_SIZE+3),VPTE_INDEX_SIZE+3+4
+	cmpldi	r10,0x80
+	bne-	virt_page_table_tlb_miss_fault
+
+	/* Get the PGD pointer */
+	ld	r15,PACAPGD(r11)
+	cmpldi	cr0,r15,0
+	beq-	virt_page_table_tlb_miss_fault
+
+	/* Get to PGD entry */
+	rldicl	r11,r16,64-VPTE_PGD_SHIFT,64-PGD_INDEX_SIZE-3
+	clrrdi	r10,r11,3
+	ldx	r15,r10,r15
+	cmpdi	cr0,r15,0
+	bge	virt_page_table_tlb_miss_fault
+
+	/* Get to PUD entry */
+	rldicl	r11,r16,64-VPTE_PUD_SHIFT,64-PUD_INDEX_SIZE-3
+	clrrdi	r10,r11,3
+	ldx	r15,r10,r15
+	cmpdi	cr0,r15,0
+	bge	virt_page_table_tlb_miss_fault
+
+	/* Get to PMD entry */
+	rldicl	r11,r16,64-VPTE_PMD_SHIFT,64-PMD_INDEX_SIZE-3
+	clrrdi	r10,r11,3
+	ldx	r15,r10,r15
+	cmpdi	cr0,r15,0
+	bge	virt_page_table_tlb_miss_fault
+
+	/* Ok, we're all right, we can now create a kernel translation for
+	 * a 4K or 64K page from r16 -> r15.
+	 */
+	/* Now we build the MAS:
+	 *
+	 * MAS 0   :	Fully setup with defaults in MAS4 and TLBnCFG
+	 * MAS 1   :	Almost fully setup
+	 *               - PID already updated by caller if necessary
+	 *               - TSIZE for now is base page size always
+	 * MAS 2   :	Use defaults
+	 * MAS 3+7 :	Needs to be done
+	 *
+	 * So we only do MAS 2 and 3 for now...
+	 */
+	clrldi	r11,r15,4		/* remove region ID from RPN */
+	ori	r10,r11,1		/* Or-in SR */
+
+	srdi	r16,r10,32
+	mtspr	SPRN_MAS3,r10
+	mtspr	SPRN_MAS7,r16
+
+	tlbwe
+
+	/* Return to caller, normal case */
+	TLB_MISS_EPILOG_SUCCESS
+	rfi
+
+virt_page_table_tlb_miss_fault:
+	/* If we fault here, things are a little bit tricky. We need to call
+	 * either data or instruction store fault, and we need to retrieve
+	 * the original fault address and ESR (for data).
+	 *
+	 * The thing is, we know that in normal circumstances, this is
+	 * always called as a second level tlb miss for SW load or as a first
+	 * level TLB miss for HW load, so we should be able to peek at the
+	 * relevant information in the first exception frame in the PACA.
+	 *
+	 * However, we do need to double check that, because we may just hit
+	 * a stray kernel pointer or a userland attack trying to hit those
+	 * areas. If that is the case, we do a data fault. (We can't get here
+	 * from an instruction tlb miss anyway).
+	 *
+	 * Note also that when going to a fault, we must unwind the previous
+	 * level as well. Since we are doing that, we don't need to clear or
+	 * restore the TLB reservation neither.
+	 */
+	subf	r10,r13,r12
+	cmpldi	cr0,r10,PACA_EXTLB+EX_TLB_SIZE
+	bne-	virt_page_table_tlb_miss_whacko_fault
+
+	/* We dig the original DEAR and ESR from slot 0 */
+	ld	r15,EX_TLB_DEAR+PACA_EXTLB(r13)
+	ld	r16,EX_TLB_ESR+PACA_EXTLB(r13)
+
+	/* We check for the "special" ESR value for instruction faults */
+	cmpdi	cr0,r16,-1
+	beq	1f
+	mtspr	SPRN_DEAR,r15
+	mtspr	SPRN_ESR,r16
+	TLB_MISS_EPILOG_ERROR
+	b	exc_data_storage_book3e
+1:	TLB_MISS_EPILOG_ERROR
+	b	exc_instruction_storage_book3e
+
+virt_page_table_tlb_miss_whacko_fault:
+	/* The linear fault will restart everything so ESR and DEAR will
+	 * not have been clobbered, let's just fault with what we have
+	 */
+	TLB_MISS_EPILOG_ERROR
+	b	exc_data_storage_book3e
+
+/*
+ * This is the guts of "any" level TLB miss handler for kernel linear
+ * mapping misses. We are entered with:
+ *
+ *
+ * r16 = faulting address
+ * r15 = crap (free to use)
+ * r14 = ESR (data) or -1 (instruction)
+ * r13 = PACA
+ * r12 = TLB exception frame in PACA
+ * r11 = crap (free to use)
+ * r10 = crap (free to use)
+ *
+ * In addition we know that we will not re-enter, so in theory, we could
+ * use a simpler epilog not restoring SRR0/1 etc.. but we'll do that later.
+ *
+ * We also need to be careful about MAS registers here & TLB reservation,
+ * as we know we'll have clobbered them if we interrupt the main TLB miss
+ * handlers in which case we probably want to do a full restart at level
+ * 0 rather than saving / restoring the MAS.
+ *
+ * Note: If we care about performance of that core, we can easily shuffle
+ *       a few things around
+ */
+tlb_load_linear:
+	/* For now, we assume the linear mapping is contiguous and stops at
+	 * linear_map_top. We also assume the size is a multiple of 1G, thus
+	 * we only use 1G pages for now. That might have to be changed in a
+	 * final implementation, especially when dealing with hypervisors
+	 */
+	__LOAD_PACA_TOC(r11)
+	LOAD_REG_ADDR_ALTTOC(r11, r11, linear_map_top)
+	ld	r10,0(r11)
+	tovirt(10,10)
+	cmpld	cr0,r16,r10
+	bge	tlb_load_linear_fault
+
+	/* MAS1 need whole new setup. */
+	li	r15,(BOOK3E_PAGESZ_1GB<<MAS1_TSIZE_SHIFT)
+	oris	r15,r15,MAS1_VALID@h	/* MAS1 needs V and TSIZE */
+	mtspr	SPRN_MAS1,r15
+
+	/* Already somebody there ? */
+	PPC_TLBSRX_DOT(0,R16)
+	beq	tlb_load_linear_done
+
+	/* Now we build the remaining MAS. MAS0 and 2 should be fine
+	 * with their defaults, which leaves us with MAS 3 and 7. The
+	 * mapping is linear, so we just take the address, clear the
+	 * region bits, and or in the permission bits which are currently
+	 * hard wired
+	 */
+	clrrdi	r10,r16,30		/* 1G page index */
+	clrldi	r10,r10,4		/* clear region bits */
+	ori	r10,r10,MAS3_SR|MAS3_SW|MAS3_SX
+
+	srdi	r16,r10,32
+	mtspr	SPRN_MAS3,r10
+	mtspr	SPRN_MAS7,r16
+
+	tlbwe
+
+tlb_load_linear_done:
+	/* We use the "error" epilog for success as we do want to
+	 * restore to the initial faulting context, whatever it was.
+	 * We do that because we can't resume a fault within a TLB
+	 * miss handler, due to MAS and TLB reservation being clobbered.
+	 */
+	TLB_MISS_EPILOG_ERROR
+	rfi
+
+tlb_load_linear_fault:
+	/* We keep the DEAR and ESR around, this shouldn't have happened */
+	cmpdi	cr0,r14,-1
+	beq	1f
+	TLB_MISS_EPILOG_ERROR_SPECIAL
+	b	exc_data_storage_book3e
+1:	TLB_MISS_EPILOG_ERROR_SPECIAL
+	b	exc_instruction_storage_book3e