[PATCH v8 07/15] ARM: KVM: Hypervisor inititalization

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Sets up the required registers to run code in HYP-mode from the kernel.

By setting the HVBAR the kernel can execute code in Hyp-mode with
the MMU disabled. The HVBAR initially points to initialization code,
which initializes other Hyp-mode registers and enables the MMU
for Hyp-mode. Afterwards, the HVBAR is changed to point to KVM
Hyp vectors used to catch guest faults and to switch to Hyp mode
to perform a world-switch into a KVM guest.

Also provides memory mapping code to map required code pages and data
structures accessed in Hyp mode at the same virtual address as the
host kernel virtual addresses, but which conforms to the architectural
requirements for translations in Hyp mode. This interface is added in
arch/arm/kvm/arm_mmu.c and is comprised of:
 - create_hyp_mappings(hyp_pgd, start, end);
 - free_hyp_pmds(pgd_hyp);

Note: The initialization mechanism currently relies on an SMC #0 call
to the secure monitor, which was merely a fast way of getting to the
hypervisor. Dave Marting and Rusty Russel have patches out to make the
boot-wrapper and the kernel boot in Hyp-mode and setup a generic way for
hypervisors to get access to Hyp-mode if the boot-loader allows such
access.

Signed-off-by: Christoffer Dall <c.dall@xxxxxxxxxxxxxxxxxxxxxx>
---
 arch/arm/include/asm/kvm_arm.h              |  117 +++++++++++++++++++
 arch/arm/include/asm/kvm_asm.h              |   22 +++
 arch/arm/include/asm/kvm_mmu.h              |   37 ++++++
 arch/arm/include/asm/pgtable-3level-hwdef.h |    4 +
 arch/arm/include/asm/pgtable-3level.h       |    4 +
 arch/arm/include/asm/pgtable.h              |    1 
 arch/arm/kvm/arm.c                          |  167 +++++++++++++++++++++++++++
 arch/arm/kvm/exports.c                      |   15 ++
 arch/arm/kvm/init.S                         |   99 ++++++++++++++++
 arch/arm/kvm/interrupts.S                   |   47 +++++++
 arch/arm/kvm/mmu.c                          |  170 +++++++++++++++++++++++++++
 mm/memory.c                                 |    2 
 12 files changed, 685 insertions(+)
 create mode 100644 arch/arm/include/asm/kvm_arm.h
 create mode 100644 arch/arm/include/asm/kvm_mmu.h

diff --git a/arch/arm/include/asm/kvm_arm.h b/arch/arm/include/asm/kvm_arm.h
new file mode 100644
index 0000000..7f30cbd
--- /dev/null
+++ b/arch/arm/include/asm/kvm_arm.h
@@ -0,0 +1,117 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ */
+
+#ifndef __KVM_ARM_H__
+#define __KVM_ARM_H__
+
+#include <asm/types.h>
+
+/* Hyp Configuration Register (HCR) bits */
+#define HCR_TGE		(1 << 27)
+#define HCR_TVM		(1 << 26)
+#define HCR_TTLB	(1 << 25)
+#define HCR_TPU		(1 << 24)
+#define HCR_TPC		(1 << 23)
+#define HCR_TSW		(1 << 22)
+#define HCR_TAC		(1 << 21)
+#define HCR_TIDCP	(1 << 20)
+#define HCR_TSC		(1 << 19)
+#define HCR_TID3	(1 << 18)
+#define HCR_TID2	(1 << 17)
+#define HCR_TID1	(1 << 16)
+#define HCR_TID0	(1 << 15)
+#define HCR_TWE		(1 << 14)
+#define HCR_TWI		(1 << 13)
+#define HCR_DC		(1 << 12)
+#define HCR_BSU		(3 << 10)
+#define HCR_BSU_IS	(1 << 10)
+#define HCR_FB		(1 << 9)
+#define HCR_VA		(1 << 8)
+#define HCR_VI		(1 << 7)
+#define HCR_VF		(1 << 6)
+#define HCR_AMO		(1 << 5)
+#define HCR_IMO		(1 << 4)
+#define HCR_FMO		(1 << 3)
+#define HCR_PTW		(1 << 2)
+#define HCR_SWIO	(1 << 1)
+#define HCR_VM		1
+
+/*
+ * The bits we set in HCR:
+ * TAC:		Trap ACTLR
+ * TSC:		Trap SMC
+ * TWI:		Trap WFI
+ * BSU_IS:	Upgrade barriers to the inner shareable domain
+ * FB:		Force broadcast of all maintainance operations
+ * AMO:		Override CPSR.A and enable signaling with VA
+ * IMO:		Override CPSR.I and enable signaling with VI
+ * FMO:		Override CPSR.F and enable signaling with VF
+ * SWIO:	Turn set/way invalidates into set/way clean+invalidate
+ */
+#define HCR_GUEST_MASK (HCR_TSC | HCR_TWI | HCR_VM | HCR_BSU_IS | HCR_FB | \
+			HCR_AMO | HCR_IMO | HCR_FMO | HCR_FMO | HCR_SWIO)
+
+/* Hyp System Control Register (HSCTLR) bits */
+#define HSCTLR_TE	(1 << 30)
+#define HSCTLR_EE	(1 << 25)
+#define HSCTLR_FI	(1 << 21)
+#define HSCTLR_WXN	(1 << 19)
+#define HSCTLR_I	(1 << 12)
+#define HSCTLR_C	(1 << 2)
+#define HSCTLR_A	(1 << 1)
+#define HSCTLR_M	1
+#define HSCTLR_MASK	(HSCTLR_M | HSCTLR_A | HSCTLR_C | HSCTLR_I | \
+			 HSCTLR_WXN | HSCTLR_FI | HSCTLR_EE | HSCTLR_TE)
+
+/* TTBCR and HTCR Registers bits */
+#define TTBCR_EAE	(1 << 31)
+#define TTBCR_IMP	(1 << 30)
+#define TTBCR_SH1	(3 << 28)
+#define TTBCR_ORGN1	(3 << 26)
+#define TTBCR_IRGN1	(3 << 24)
+#define TTBCR_EPD1	(1 << 23)
+#define TTBCR_A1	(1 << 22)
+#define TTBCR_T1SZ	(3 << 16)
+#define TTBCR_SH0	(3 << 12)
+#define TTBCR_ORGN0	(3 << 10)
+#define TTBCR_IRGN0	(3 << 8)
+#define TTBCR_EPD0	(1 << 7)
+#define TTBCR_T0SZ	3
+#define HTCR_MASK	(TTBCR_T0SZ | TTBCR_IRGN0 | TTBCR_ORGN0 | TTBCR_SH0)
+
+
+/* Virtualization Translation Control Register (VTCR) bits */
+#define VTCR_SH0	(3 << 12)
+#define VTCR_ORGN0	(3 << 10)
+#define VTCR_IRGN0	(3 << 8)
+#define VTCR_SL0	(3 << 6)
+#define VTCR_S		(1 << 4)
+#define VTCR_T0SZ	3
+#define VTCR_MASK	(VTCR_SH0 | VTCR_ORGN0 | VTCR_IRGN0 | VTCR_SL0 | \
+			 VTCR_S | VTCR_T0SZ | VTCR_MASK)
+#define VTCR_HTCR_SH	(VTCR_SH0 | VTCR_ORGN0 | VTCR_IRGN0)
+#define VTCR_SL_L2	0		/* Starting-level: 2 */
+#define VTCR_SL_L1	(1 << 6)	/* Starting-level: 1 */
+#define VTCR_GUEST_SL	VTCR_SL_L1
+#define VTCR_GUEST_T0SZ	0
+#if VTCR_GUEST_SL == 0
+#define VTTBR_X		(14 - VTCR_GUEST_T0SZ)
+#else
+#define VTTBR_X		(5 - VTCR_GUEST_T0SZ)
+#endif
+
+
+#endif /* __KVM_ARM_H__ */
diff --git a/arch/arm/include/asm/kvm_asm.h b/arch/arm/include/asm/kvm_asm.h
index c3d4458..69afdf3 100644
--- a/arch/arm/include/asm/kvm_asm.h
+++ b/arch/arm/include/asm/kvm_asm.h
@@ -24,5 +24,27 @@
 #define ARM_EXCEPTION_DATA_ABORT  4
 #define ARM_EXCEPTION_IRQ	  5
 #define ARM_EXCEPTION_FIQ	  6
+#define ARM_EXCEPTION_HVC	  7
+
+/*
+ * SMC Hypervisor API call number
+ */
+#define SMCHYP_HVBAR_W 0xfffffff0
+
+#ifndef __ASSEMBLY__
+struct kvm_vcpu;
+
+extern char __kvm_hyp_init[];
+extern char __kvm_hyp_init_end[];
+
+extern char __kvm_hyp_vector[];
+
+extern char __kvm_hyp_code_start[];
+extern char __kvm_hyp_code_end[];
+
+extern void __kvm_flush_vm_context(void);
+
+extern int __kvm_vcpu_run(struct kvm_vcpu *vcpu);
+#endif
 
 #endif /* __ARM_KVM_ASM_H__ */
diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h
new file mode 100644
index 0000000..1aa1af4
--- /dev/null
+++ b/arch/arm/include/asm/kvm_mmu.h
@@ -0,0 +1,37 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
+ *
+ */
+
+#ifndef __ARM_KVM_MMU_H__
+#define __ARM_KVM_MMU_H__
+
+/*
+ * The architecture supports 40-bit IPA as input to the 2nd stage translations
+ * and PTRS_PER_PGD2 could therefore be 1024.
+ *
+ * To save a bit of memory and to avoid alignment issues we assume 39-bit IPA
+ * for now, but remember that the level-1 table must be aligned to its size.
+ */
+#define PTRS_PER_PGD2	512
+#define PGD2_ORDER	get_order(PTRS_PER_PGD2 * sizeof(pgd_t))
+
+int create_hyp_mappings(void *from, void *to);
+void free_hyp_pmds(void);
+
+int kvm_hyp_pgd_alloc(void);
+pgd_t *kvm_hyp_pgd_get(void);
+void kvm_hyp_pgd_free(void);
+
+#endif /* __ARM_KVM_MMU_H__ */
diff --git a/arch/arm/include/asm/pgtable-3level-hwdef.h b/arch/arm/include/asm/pgtable-3level-hwdef.h
index a2d404e..18f5cef 100644
--- a/arch/arm/include/asm/pgtable-3level-hwdef.h
+++ b/arch/arm/include/asm/pgtable-3level-hwdef.h
@@ -32,6 +32,9 @@
 #define PMD_TYPE_SECT		(_AT(pmdval_t, 1) << 0)
 #define PMD_BIT4		(_AT(pmdval_t, 0))
 #define PMD_DOMAIN(x)		(_AT(pmdval_t, 0))
+#define PMD_APTABLE_SHIFT	(61)
+#define PMD_APTABLE		(_AT(pgdval_t, 3) << PGD_APTABLE_SHIFT)
+#define PMD_PXNTABLE		(_AT(pgdval_t, 1) << 59)
 
 /*
  *   - section
@@ -41,6 +44,7 @@
 #define PMD_SECT_S		(_AT(pmdval_t, 3) << 8)
 #define PMD_SECT_AF		(_AT(pmdval_t, 1) << 10)
 #define PMD_SECT_nG		(_AT(pmdval_t, 1) << 11)
+#define PMD_SECT_PXN		(_AT(pmdval_t, 1) << 53)
 #define PMD_SECT_XN		(_AT(pmdval_t, 1) << 54)
 #define PMD_SECT_AP_WRITE	(_AT(pmdval_t, 0))
 #define PMD_SECT_AP_READ	(_AT(pmdval_t, 0))
diff --git a/arch/arm/include/asm/pgtable-3level.h b/arch/arm/include/asm/pgtable-3level.h
index b249035..1169a8a 100644
--- a/arch/arm/include/asm/pgtable-3level.h
+++ b/arch/arm/include/asm/pgtable-3level.h
@@ -107,6 +107,10 @@
 #define pud_none(pud)		(!pud_val(pud))
 #define pud_bad(pud)		(!(pud_val(pud) & 2))
 #define pud_present(pud)	(pud_val(pud))
+#define pmd_table(pmd)		((pmd_val(pmd) & PMD_TYPE_MASK) == \
+						 PMD_TYPE_TABLE)
+#define pmd_sect(pmd)		((pmd_val(pmd) & PMD_TYPE_MASK) == \
+						 PMD_TYPE_SECT)
 
 #define pud_clear(pudp)			\
 	do {				\
diff --git a/arch/arm/include/asm/pgtable.h b/arch/arm/include/asm/pgtable.h
index c7bd809..4b72287 100644
--- a/arch/arm/include/asm/pgtable.h
+++ b/arch/arm/include/asm/pgtable.h
@@ -82,6 +82,7 @@ extern pgprot_t		pgprot_kernel;
 #define PAGE_READONLY_EXEC	_MOD_PROT(pgprot_user, L_PTE_USER | L_PTE_RDONLY)
 #define PAGE_KERNEL		_MOD_PROT(pgprot_kernel, L_PTE_XN)
 #define PAGE_KERNEL_EXEC	pgprot_kernel
+#define PAGE_HYP		_MOD_PROT(pgprot_kernel, L_PTE_USER)
 
 #define __PAGE_NONE		__pgprot(_L_PTE_DEFAULT | L_PTE_RDONLY | L_PTE_XN)
 #define __PAGE_SHARED		__pgprot(_L_PTE_DEFAULT | L_PTE_USER | L_PTE_XN)
diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
index 5992d90..4c61d3c 100644
--- a/arch/arm/kvm/arm.c
+++ b/arch/arm/kvm/arm.c
@@ -31,6 +31,12 @@
 #include <asm/uaccess.h>
 #include <asm/ptrace.h>
 #include <asm/mman.h>
+#include <asm/tlbflush.h>
+#include <asm/kvm_arm.h>
+#include <asm/kvm_asm.h>
+#include <asm/kvm_mmu.h>
+
+static DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page);
 
 int kvm_arch_hardware_enable(void *garbage)
 {
@@ -255,13 +261,174 @@ long kvm_arch_vm_ioctl(struct file *filp,
 	return -EINVAL;
 }
 
+static void cpu_set_vector(void *vector)
+{
+	unsigned long vector_ptr;
+	unsigned long smc_hyp_nr;
+
+	vector_ptr = (unsigned long)vector;
+	smc_hyp_nr = SMCHYP_HVBAR_W;
+
+	/*
+	 * Set the HVBAR
+	 */
+	asm volatile (
+		"mov	r0, %[vector_ptr]\n\t"
+		"mov	r7, %[smc_hyp_nr]\n\t"
+		"smc	#0\n\t" : :
+		[vector_ptr] "r" (vector_ptr),
+		[smc_hyp_nr] "r" (smc_hyp_nr) :
+		"r0", "r7");
+}
+
+static void cpu_init_hyp_mode(void *vector)
+{
+	unsigned long pgd_ptr;
+	unsigned long hyp_stack_ptr;
+	unsigned long stack_page;
+
+	cpu_set_vector(vector);
+
+	pgd_ptr = virt_to_phys(kvm_hyp_pgd_get());
+	stack_page = __get_cpu_var(kvm_arm_hyp_stack_page);
+	hyp_stack_ptr = stack_page + PAGE_SIZE;
+
+	/*
+	 * Call initialization code
+	 */
+	asm volatile (
+		"mov	r0, %[pgd_ptr]\n\t"
+		"mov	r1, %[hyp_stack_ptr]\n\t"
+		"hvc	#0\n\t" : :
+		[pgd_ptr] "r" (pgd_ptr),
+		[hyp_stack_ptr] "r" (hyp_stack_ptr) :
+		"r0", "r1");
+}
+
+/**
+ * Inits Hyp-mode on all online CPUs
+ */
+static int init_hyp_mode(void)
+{
+	phys_addr_t init_phys_addr, init_end_phys_addr;
+	int cpu;
+	int err = 0;
+
+	/*
+	 * Allocate stack pages for Hypervisor-mode
+	 */
+	for_each_possible_cpu(cpu) {
+		unsigned long stack_page;
+
+		stack_page = __get_free_page(GFP_KERNEL);
+		if (!stack_page) {
+			err = -ENOMEM;
+			goto out_free_stack_pages;
+		}
+
+		per_cpu(kvm_arm_hyp_stack_page, cpu) = stack_page;
+	}
+
+	/*
+	 * Allocate Hyp level-1 page table
+	 */
+	err = kvm_hyp_pgd_alloc();
+	if (err)
+		goto out_free_stack_pages;
+
+	init_phys_addr = virt_to_phys(__kvm_hyp_init);
+	init_end_phys_addr = virt_to_phys(__kvm_hyp_init_end);
+	BUG_ON(init_phys_addr & 0x1f);
+
+	/*
+	 * Create identity mapping for the init code.
+	 */
+	hyp_idmap_add(kvm_hyp_pgd_get(),
+		      (unsigned long)init_phys_addr,
+		      (unsigned long)init_end_phys_addr);
+
+	/*
+	 * Execute the init code on each CPU.
+	 *
+	 * Note: The stack is not mapped yet, so don't do anything else than
+	 * initializing the hypervisor mode on each CPU using a local stack
+	 * space for temporary storage.
+	 */
+	for_each_online_cpu(cpu) {
+		smp_call_function_single(cpu, cpu_init_hyp_mode,
+					 (void *)(long)init_phys_addr, 1);
+	}
+
+	/*
+	 * Unmap the identity mapping
+	 */
+	hyp_idmap_del(kvm_hyp_pgd_get(),
+		      (unsigned long)init_phys_addr,
+		      (unsigned long)init_end_phys_addr);
+
+	/*
+	 * Map the Hyp-code called directly from the host
+	 */
+	err = create_hyp_mappings(__kvm_hyp_code_start, __kvm_hyp_code_end);
+	if (err) {
+		kvm_err("Cannot map world-switch code\n");
+		goto out_free_mappings;
+	}
+
+	/*
+	 * Map the Hyp stack pages
+	 */
+	for_each_possible_cpu(cpu) {
+		char *stack_page = (char *)per_cpu(kvm_arm_hyp_stack_page, cpu);
+		err = create_hyp_mappings(stack_page, stack_page + PAGE_SIZE);
+
+		if (err) {
+			kvm_err("Cannot map hyp stack\n");
+			goto out_free_mappings;
+		}
+	}
+
+	/*
+	 * Set the HVBAR to the virtual kernel address
+	 */
+	for_each_online_cpu(cpu)
+		smp_call_function_single(cpu, cpu_set_vector,
+					 __kvm_hyp_vector, 1);
+
+	return 0;
+out_free_mappings:
+	free_hyp_pmds();
+	kvm_hyp_pgd_free();
+out_free_stack_pages:
+	for_each_possible_cpu(cpu)
+		free_page(per_cpu(kvm_arm_hyp_stack_page, cpu));
+	return err;
+}
+
+/**
+ * Initialize Hyp-mode and memory mappings on all CPUs.
+ */
 int kvm_arch_init(void *opaque)
 {
+	int err;
+
+	err = init_hyp_mode();
+	if (err)
+		goto out_err;
+
 	return 0;
+out_err:
+	return err;
 }
 
 void kvm_arch_exit(void)
 {
+	int cpu;
+
+	free_hyp_pmds();
+	for_each_possible_cpu(cpu)
+		free_page(per_cpu(kvm_arm_hyp_stack_page, cpu));
+	kvm_hyp_pgd_free();
 }
 
 static int arm_init(void)
diff --git a/arch/arm/kvm/exports.c b/arch/arm/kvm/exports.c
index d8a7fd5..2631609 100644
--- a/arch/arm/kvm/exports.c
+++ b/arch/arm/kvm/exports.c
@@ -14,3 +14,18 @@
  */
 
 #include <linux/module.h>
+#include <asm/kvm_asm.h>
+
+EXPORT_SYMBOL_GPL(__kvm_hyp_init);
+EXPORT_SYMBOL_GPL(__kvm_hyp_init_end);
+
+EXPORT_SYMBOL_GPL(__kvm_hyp_vector);
+
+EXPORT_SYMBOL_GPL(__kvm_hyp_code_start);
+EXPORT_SYMBOL_GPL(__kvm_hyp_code_end);
+
+EXPORT_SYMBOL_GPL(__kvm_vcpu_run);
+
+EXPORT_SYMBOL_GPL(__kvm_flush_vm_context);
+
+EXPORT_SYMBOL_GPL(smp_send_reschedule);
diff --git a/arch/arm/kvm/init.S b/arch/arm/kvm/init.S
index 073a494..7800023 100644
--- a/arch/arm/kvm/init.S
+++ b/arch/arm/kvm/init.S
@@ -13,5 +13,104 @@
  * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
  */
+
+#include <linux/linkage.h>
+#include <asm/unified.h>
 #include <asm/asm-offsets.h>
 #include <asm/kvm_asm.h>
+#include <asm/kvm_arm.h>
+
+@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+@  Hypervisor initialization
+@    - should be called with:
+@        r0 = Hypervisor pgd pointer
+@        r1 = top of Hyp stack (kernel VA)
+@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+	.text
+	.arm
+	.align 12
+__kvm_hyp_init:
+	.globl __kvm_hyp_init
+
+	@ Hyp-mode exception vector
+	nop
+	nop
+	nop
+	nop
+	nop
+	b	__do_hyp_init
+	nop
+	nop
+
+__do_hyp_init:
+	@ Set the sp to end of this page and push data for later use
+	mov	sp, pc
+	bic	sp, sp, #0x0ff
+	bic	sp, sp, #0xf00
+	add	sp, sp, #0x1000
+	push	{r0, r1, r2, r12}
+
+	@ Set the HTTBR to point to the hypervisor PGD pointer passed to
+	@ function and set the upper bits equal to the kernel PGD.
+	mrrc	p15, 1, r1, r2, c2
+	mcrr	p15, 4, r0, r2, c2
+
+	@ Set the HTCR and VTCR to the same shareability and cacheability
+	@ settings as the non-secure TTBCR and with T0SZ == 0.
+	mrc	p15, 4, r0, c2, c0, 2	@ HTCR
+	ldr	r12, =HTCR_MASK
+	bic	r0, r0, r12
+	mrc	p15, 0, r1, c2, c0, 2	@ TTBCR
+	and	r1, r1, #(HTCR_MASK & ~TTBCR_T0SZ)
+	orr	r0, r0, r1
+	mcr	p15, 4, r0, c2, c0, 2	@ HTCR
+
+	mrc	p15, 4, r1, c2, c1, 2	@ VTCR
+	bic	r1, r1, #(VTCR_HTCR_SH | VTCR_SL0)
+	bic	r0, r0, #(~VTCR_HTCR_SH)
+	orr	r1, r0, r1
+	orr	r1, r1, #(VTCR_SL_L1 | VTCR_GUEST_T0SZ)
+	mcr	p15, 4, r1, c2, c1, 2	@ VTCR
+
+	@ Use the same memory attributes for hyp. accesses as the kernel
+	@ (copy MAIRx ro HMAIRx).
+	mrc	p15, 0, r0, c10, c2, 0
+	mcr	p15, 4, r0, c10, c2, 0
+	mrc	p15, 0, r0, c10, c2, 1
+	mcr	p15, 4, r0, c10, c2, 1
+
+	@ Set the HSCTLR to:
+	@  - ARM/THUMB exceptions: Kernel config (Thumb-2 kernel)
+	@  - Endianness: Kernel config
+	@  - Fast Interrupt Features: Kernel config
+	@  - Write permission implies XN: disabled
+	@  - Instruction cache: enabled
+	@  - Data/Unified cache: enabled
+	@  - Memory alignment checks: enabled
+	@  - MMU: enabled (this code must be run from an identity mapping)
+	mrc	p15, 4, r0, c1, c0, 0	@ HSCR
+	ldr	r12, =HSCTLR_MASK
+	bic	r0, r0, r12
+	mrc	p15, 0, r1, c1, c0, 0	@ SCTLR
+	ldr	r12, =(HSCTLR_EE | HSCTLR_FI)
+	and	r1, r1, r12
+ ARM(	ldr	r12, =(HSCTLR_M | HSCTLR_A | HSCTLR_I)			)
+ THUMB(	ldr	r12, =(HSCTLR_M | HSCTLR_A | HSCTLR_I | HSCTLR_TE)	)
+	orr	r1, r1, r12
+	orr	r0, r0, r1
+	isb
+	mcr	p15, 4, r0, c1, c0, 0	@ HSCR
+	isb
+
+	@ Set stack pointer and return to the kernel
+	pop	{r0, r1, r2, r12}
+	mov	sp, r1
+	eret
+
+	.ltorg
+
+	.align 12
+
+	__kvm_init_sp:
+	.globl __kvm_hyp_init_end
+__kvm_hyp_init_end:
diff --git a/arch/arm/kvm/interrupts.S b/arch/arm/kvm/interrupts.S
index 073a494..d99a9c7 100644
--- a/arch/arm/kvm/interrupts.S
+++ b/arch/arm/kvm/interrupts.S
@@ -13,5 +13,52 @@
  * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
  */
+
+#include <linux/linkage.h>
+#include <linux/const.h>
+#include <asm/unified.h>
+#include <asm/page.h>
 #include <asm/asm-offsets.h>
 #include <asm/kvm_asm.h>
+#include <asm/kvm_arm.h>
+
+	.text
+	.align	PAGE_SHIFT
+
+__kvm_hyp_code_start:
+	.globl __kvm_hyp_code_start
+
+@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+@  Flush TLBs and instruction caches of current CPU for all VMIDs
+@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+
+ENTRY(__kvm_flush_vm_context)
+	bx	lr
+
+@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+@  Hypervisor world-switch code
+@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+
+ENTRY(__kvm_vcpu_run)
+	bx	lr
+
+
+@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+@  Hypervisor exception vector and handlers
+@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+
+	.align 5
+__kvm_hyp_vector:
+	.globl __kvm_hyp_vector
+	nop
+
+/*
+ * The below lines makes sure the HYP mode code fits in a single page (the
+ * assembler will bark at you if it doesn't). Please keep them together. If
+ * you plan to restructure the code or increase its size over a page, you'll
+ * have to fix the code in init_hyp_mode().
+ */
+__kvm_hyp_code_end:
+	.globl	__kvm_hyp_code_end
+
+	.org	__kvm_hyp_code_start + PAGE_SIZE
diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c
index 2cccd48..a320b56a 100644
--- a/arch/arm/kvm/mmu.c
+++ b/arch/arm/kvm/mmu.c
@@ -13,3 +13,173 @@
  * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  *
  */
+
+#include <linux/mman.h>
+#include <linux/kvm_host.h>
+#include <asm/pgalloc.h>
+#include <asm/kvm_arm.h>
+#include <asm/kvm_mmu.h>
+
+static pgd_t *kvm_hyp_pgd;
+static DEFINE_MUTEX(kvm_hyp_pgd_mutex);
+
+static void free_ptes(pmd_t *pmd, unsigned long addr)
+{
+	pte_t *pte;
+	unsigned int i;
+
+	for (i = 0; i < PTRS_PER_PMD; i++, addr += PMD_SIZE) {
+		if (!pmd_none(*pmd) && pmd_table(*pmd)) {
+			pte = pte_offset_kernel(pmd, addr);
+			pte_free_kernel(NULL, pte);
+		}
+		pmd++;
+	}
+}
+
+/**
+ * free_hyp_pmds - free a Hyp-mode level-2 tables and child level-3 tables
+ *
+ * Assumes this is a page table used strictly in Hyp-mode and therefore contains
+ * only mappings in the kernel memory area, which is above PAGE_OFFSET.
+ */
+void free_hyp_pmds(void)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	unsigned long addr;
+
+	mutex_lock(&kvm_hyp_pgd_mutex);
+	for (addr = PAGE_OFFSET; addr != 0; addr += PGDIR_SIZE) {
+		pgd = kvm_hyp_pgd + pgd_index(addr);
+		pud = pud_offset(pgd, addr);
+
+		if (pud_none(*pud))
+			continue;
+		BUG_ON(pud_bad(*pud));
+
+		pmd = pmd_offset(pud, addr);
+		free_ptes(pmd, addr);
+		pmd_free(NULL, pmd);
+	}
+	mutex_unlock(&kvm_hyp_pgd_mutex);
+}
+
+static void create_hyp_pte_mappings(pmd_t *pmd, unsigned long start,
+						unsigned long end)
+{
+	pte_t *pte;
+	struct page *page;
+	unsigned long addr;
+
+	for (addr = start & PAGE_MASK; addr < end; addr += PAGE_SIZE) {
+		pte = pte_offset_kernel(pmd, addr);
+		BUG_ON(!virt_addr_valid(addr));
+		page = virt_to_page(addr);
+
+		set_pte_ext(pte, mk_pte(page, PAGE_HYP), 0);
+	}
+}
+
+static int create_hyp_pmd_mappings(pud_t *pud, unsigned long start,
+					       unsigned long end)
+{
+	pmd_t *pmd;
+	pte_t *pte;
+	unsigned long addr, next;
+
+	for (addr = start; addr < end; addr = next) {
+		pmd = pmd_offset(pud, addr);
+
+		BUG_ON(pmd_sect(*pmd));
+
+		if (pmd_none(*pmd)) {
+			pte = pte_alloc_one_kernel(NULL, addr);
+			if (!pte) {
+				kvm_err("Cannot allocate Hyp pte\n");
+				return -ENOMEM;
+			}
+			pmd_populate_kernel(NULL, pmd, pte);
+		}
+
+		next = pmd_addr_end(addr, end);
+		create_hyp_pte_mappings(pmd, addr, next);
+	}
+
+	return 0;
+}
+
+/**
+ * create_hyp_mappings - map a kernel virtual address range in Hyp mode
+ * @from:	The virtual kernel start address of the range
+ * @to:		The virtual kernel end address of the range (exclusive)
+ *
+ * The same virtual address as the kernel virtual address is also used in
+ * Hyp-mode mapping to the same underlying physical pages.
+ *
+ * Note: Wrapping around zero in the "to" address is not supported.
+ */
+int create_hyp_mappings(void *from, void *to)
+{
+	unsigned long start = (unsigned long)from;
+	unsigned long end = (unsigned long)to;
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	unsigned long addr, next;
+	int err = 0;
+
+	BUG_ON(start > end);
+	if (start < PAGE_OFFSET)
+		return -EINVAL;
+
+	mutex_lock(&kvm_hyp_pgd_mutex);
+	for (addr = start; addr < end; addr = next) {
+		pgd = kvm_hyp_pgd + pgd_index(addr);
+		pud = pud_offset(pgd, addr);
+
+		if (pud_none_or_clear_bad(pud)) {
+			pmd = pmd_alloc_one(NULL, addr);
+			if (!pmd) {
+				kvm_err("Cannot allocate Hyp pmd\n");
+				err = -ENOMEM;
+				goto out;
+			}
+			pud_populate(NULL, pud, pmd);
+		}
+
+		next = pgd_addr_end(addr, end);
+		err = create_hyp_pmd_mappings(pud, addr, next);
+		if (err)
+			goto out;
+	}
+out:
+	mutex_unlock(&kvm_hyp_pgd_mutex);
+	return err;
+}
+
+int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
+{
+	return -EINVAL;
+}
+
+int kvm_hyp_pgd_alloc(void)
+{
+	kvm_hyp_pgd = kzalloc(PTRS_PER_PGD * sizeof(pgd_t), GFP_KERNEL);
+	if (!kvm_hyp_pgd)
+		return -ENOMEM;
+
+	return 0;
+}
+
+pgd_t *kvm_hyp_pgd_get(void)
+{
+	return kvm_hyp_pgd;
+}
+
+void kvm_hyp_pgd_free(void)
+{
+	kfree(kvm_hyp_pgd);
+	kvm_hyp_pgd = NULL;
+}
diff --git a/mm/memory.c b/mm/memory.c
index 1b7dc66..b09d781 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -379,12 +379,14 @@ void pgd_clear_bad(pgd_t *pgd)
 	pgd_ERROR(*pgd);
 	pgd_clear(pgd);
 }
+EXPORT_SYMBOL_GPL(pgd_clear_bad);
 
 void pud_clear_bad(pud_t *pud)
 {
 	pud_ERROR(*pud);
 	pud_clear(pud);
 }
+EXPORT_SYMBOL_GPL(pud_clear_bad);
 
 void pmd_clear_bad(pmd_t *pmd)
 {

--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]
  Powered by Linux