X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=arch%2Fx86%2Fkernel%2Fparavirt.c;h=1b1739d1631028edbf672666043de00cdc94aa6d;hb=918bc960dc630b1a79c0d2991a81985812ff69f5;hp=1a170877f46cb35442d2ea5198be1938ab695da5;hpb=b1df07bd6674a84fbd9248759dc3fa3ff5c78e5b;p=safe%2Fjmp%2Flinux-2.6

diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index 1a17087..1b1739d 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -28,8 +28,9 @@
 #include <asm/paravirt.h>
 #include <asm/desc.h>
 #include <asm/setup.h>
-#include <asm/arch_hooks.h>
+#include <asm/pgtable.h>
 #include <asm/time.h>
+#include <asm/pgalloc.h>
 #include <asm/irq.h>
 #include <asm/delay.h>
 #include <asm/fixmap.h>
@@ -42,15 +43,21 @@ void _paravirt_nop(void)
 {
 }
 
-static void __init default_banner(void)
+/* identity function, which can be inlined */
+u32 _paravirt_ident_32(u32 x)
 {
-	printk(KERN_INFO "Booting paravirtualized kernel on %s\n",
-	       pv_info.name);
+	return x;
 }
 
-char *memory_setup(void)
+u64 _paravirt_ident_64(u64 x)
 {
-	return pv_init_ops.memory_setup();
+	return x;
+}
+
+void __init default_banner(void)
+{
+	printk(KERN_INFO "Booting paravirtualized kernel on %s\n",
+	       pv_info.name);
 }
 
 /* Simple instruction patching code. */
@@ -58,59 +65,9 @@ char *memory_setup(void)
 	extern const char start_##ops##_##name[], end_##ops##_##name[];	\
 	asm("start_" #ops "_" #name ": " code "; end_" #ops "_" #name ":")
 
-DEF_NATIVE(pv_irq_ops, irq_disable, "cli");
-DEF_NATIVE(pv_irq_ops, irq_enable, "sti");
-DEF_NATIVE(pv_irq_ops, restore_fl, "push %eax; popf");
-DEF_NATIVE(pv_irq_ops, save_fl, "pushf; pop %eax");
-DEF_NATIVE(pv_cpu_ops, iret, "iret");
-DEF_NATIVE(pv_cpu_ops, irq_enable_syscall_ret, "sti; sysexit");
-DEF_NATIVE(pv_mmu_ops, read_cr2, "mov %cr2, %eax");
-DEF_NATIVE(pv_mmu_ops, write_cr3, "mov %eax, %cr3");
-DEF_NATIVE(pv_mmu_ops, read_cr3, "mov %cr3, %eax");
-DEF_NATIVE(pv_cpu_ops, clts, "clts");
-DEF_NATIVE(pv_cpu_ops, read_tsc, "rdtsc");
-
 /* Undefined instruction for dealing with missing ops pointers. */
 static const unsigned char ud2a[] = { 0x0f, 0x0b };
 
-static unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
-			     unsigned long addr, unsigned len)
-{
-	const unsigned char *start, *end;
-	unsigned ret;
-
-	switch(type) {
-#define SITE(ops, x)						\
-	case PARAVIRT_PATCH(ops.x):				\
-		start = start_##ops##_##x;			\
-		end = end_##ops##_##x;				\
-		goto patch_site
-
-	SITE(pv_irq_ops, irq_disable);
-	SITE(pv_irq_ops, irq_enable);
-	SITE(pv_irq_ops, restore_fl);
-	SITE(pv_irq_ops, save_fl);
-	SITE(pv_cpu_ops, iret);
-	SITE(pv_cpu_ops, irq_enable_syscall_ret);
-	SITE(pv_mmu_ops, read_cr2);
-	SITE(pv_mmu_ops, read_cr3);
-	SITE(pv_mmu_ops, write_cr3);
-	SITE(pv_cpu_ops, clts);
-	SITE(pv_cpu_ops, read_tsc);
-#undef SITE
-
-	patch_site:
-		ret = paravirt_patch_insns(ibuf, len, start, end);
-		break;
-
-	default:
-		ret = paravirt_patch_default(type, clobbers, ibuf, addr, len);
-		break;
-	}
-
-	return ret;
-}
-
 unsigned paravirt_patch_nop(void)
 {
 	return 0;
@@ -172,6 +129,9 @@ static void *get_call_destination(u8 type)
 		.pv_irq_ops = pv_irq_ops,
 		.pv_apic_ops = pv_apic_ops,
 		.pv_mmu_ops = pv_mmu_ops,
+#ifdef CONFIG_PARAVIRT_SPINLOCKS
+		.pv_lock_ops = pv_lock_ops,
+#endif
 	};
 	return *((void **)&tmpl + type);
 }
@@ -185,11 +145,20 @@ unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf,
 	if (opfunc == NULL)
 		/* If there's no function, patch it with a ud2a (BUG) */
 		ret = paravirt_patch_insns(insnbuf, len, ud2a, ud2a+sizeof(ud2a));
-	else if (opfunc == paravirt_nop)
+	else if (opfunc == _paravirt_nop)
 		/* If the operation is a nop, then nop the callsite */
 		ret = paravirt_patch_nop();
+
+	/* identity functions just return their single argument */
+	else if (opfunc == _paravirt_ident_32)
+		ret = paravirt_patch_ident_32(insnbuf, len);
+	else if (opfunc == _paravirt_ident_64)
+		ret = paravirt_patch_ident_64(insnbuf, len);
+
 	else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) ||
-		 type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_syscall_ret))
+		 type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit) ||
+		 type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret32) ||
+		 type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret64))
 		/* If operation requires a jmp, then jmp */
 		ret = paravirt_patch_jmp(insnbuf, opfunc, addr, len);
 	else
@@ -214,11 +183,6 @@ unsigned paravirt_patch_insns(void *insnbuf, unsigned len,
 	return insn_len;
 }
 
-void init_IRQ(void)
-{
-	pv_irq_ops.init_IRQ();
-}
-
 static void native_flush_tlb(void)
 {
 	__native_flush_tlb();
@@ -240,14 +204,9 @@ static void native_flush_tlb_single(unsigned long addr)
 
 /* These are in entry.S */
 extern void native_iret(void);
-extern void native_irq_enable_syscall_ret(void);
-
-static int __init print_banner(void)
-{
-	pv_init_ops.banner();
-	return 0;
-}
-core_initcall(print_banner);
+extern void native_irq_enable_sysexit(void);
+extern void native_usergs_sysret32(void);
+extern void native_usergs_sysret64(void);
 
 static struct resource reserve_ioports = {
 	.start = 0,
@@ -256,13 +215,6 @@ static struct resource reserve_ioports = {
 	.flags = IORESOURCE_IO | IORESOURCE_BUSY,
 };
 
-static struct resource reserve_iomem = {
-	.start = 0,
-	.end = -1,
-	.name = "paravirt-iomem",
-	.flags = IORESOURCE_MEM | IORESOURCE_BUSY,
-};
-
 /*
  * Reserve the whole legacy IO space to prevent any legacy drivers
  * from wasting time probing for their hardware.  This is a fairly
@@ -272,34 +224,23 @@ static struct resource reserve_iomem = {
  */
 int paravirt_disable_iospace(void)
 {
-	int ret;
-
-	ret = request_resource(&ioport_resource, &reserve_ioports);
-	if (ret == 0) {
-		ret = request_resource(&iomem_resource, &reserve_iomem);
-		if (ret)
-			release_resource(&reserve_ioports);
-	}
-
-	return ret;
+	return request_resource(&ioport_resource, &reserve_ioports);
 }
 
 static DEFINE_PER_CPU(enum paravirt_lazy_mode, paravirt_lazy_mode) = PARAVIRT_LAZY_NONE;
 
 static inline void enter_lazy(enum paravirt_lazy_mode mode)
 {
-	BUG_ON(x86_read_percpu(paravirt_lazy_mode) != PARAVIRT_LAZY_NONE);
-	BUG_ON(preemptible());
+	BUG_ON(percpu_read(paravirt_lazy_mode) != PARAVIRT_LAZY_NONE);
 
-	x86_write_percpu(paravirt_lazy_mode, mode);
+	percpu_write(paravirt_lazy_mode, mode);
 }
 
-void paravirt_leave_lazy(enum paravirt_lazy_mode mode)
+static void leave_lazy(enum paravirt_lazy_mode mode)
 {
-	BUG_ON(x86_read_percpu(paravirt_lazy_mode) != mode);
-	BUG_ON(preemptible());
+	BUG_ON(percpu_read(paravirt_lazy_mode) != mode);
 
-	x86_write_percpu(paravirt_lazy_mode, PARAVIRT_LAZY_NONE);
+	percpu_write(paravirt_lazy_mode, PARAVIRT_LAZY_NONE);
 }
 
 void paravirt_enter_lazy_mmu(void)
@@ -309,22 +250,48 @@ void paravirt_enter_lazy_mmu(void)
 
 void paravirt_leave_lazy_mmu(void)
 {
-	paravirt_leave_lazy(PARAVIRT_LAZY_MMU);
+	leave_lazy(PARAVIRT_LAZY_MMU);
 }
 
-void paravirt_enter_lazy_cpu(void)
+void paravirt_start_context_switch(struct task_struct *prev)
 {
+	BUG_ON(preemptible());
+
+	if (percpu_read(paravirt_lazy_mode) == PARAVIRT_LAZY_MMU) {
+		arch_leave_lazy_mmu_mode();
+		set_ti_thread_flag(task_thread_info(prev), TIF_LAZY_MMU_UPDATES);
+	}
 	enter_lazy(PARAVIRT_LAZY_CPU);
 }
 
-void paravirt_leave_lazy_cpu(void)
+void paravirt_end_context_switch(struct task_struct *next)
 {
-	paravirt_leave_lazy(PARAVIRT_LAZY_CPU);
+	BUG_ON(preemptible());
+
+	leave_lazy(PARAVIRT_LAZY_CPU);
+
+	if (test_and_clear_ti_thread_flag(task_thread_info(next), TIF_LAZY_MMU_UPDATES))
+		arch_enter_lazy_mmu_mode();
 }
 
 enum paravirt_lazy_mode paravirt_get_lazy_mode(void)
 {
-	return x86_read_percpu(paravirt_lazy_mode);
+	if (in_interrupt())
+		return PARAVIRT_LAZY_NONE;
+
+	return percpu_read(paravirt_lazy_mode);
+}
+
+void arch_flush_lazy_mmu_mode(void)
+{
+	preempt_disable();
+
+	if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) {
+		arch_leave_lazy_mmu_mode();
+		arch_enter_lazy_mmu_mode();
+	}
+
+	preempt_enable();
 }
 
 struct pv_info pv_info = {
@@ -336,27 +303,22 @@ struct pv_info pv_info = {
 
 struct pv_init_ops pv_init_ops = {
 	.patch = native_patch,
-	.banner = default_banner,
-	.arch_setup = paravirt_nop,
-	.memory_setup = machine_specific_memory_setup,
 };
 
 struct pv_time_ops pv_time_ops = {
-	.time_init = hpet_time_init,
-	.get_wallclock = native_get_wallclock,
-	.set_wallclock = native_set_wallclock,
 	.sched_clock = native_sched_clock,
-	.get_cpu_khz = native_calculate_cpu_khz,
 };
 
 struct pv_irq_ops pv_irq_ops = {
-	.init_IRQ = native_init_IRQ,
-	.save_fl = native_save_fl,
-	.restore_fl = native_restore_fl,
-	.irq_disable = native_irq_disable,
-	.irq_enable = native_irq_enable,
+	.save_fl = __PV_IS_CALLEE_SAVE(native_save_fl),
+	.restore_fl = __PV_IS_CALLEE_SAVE(native_restore_fl),
+	.irq_disable = __PV_IS_CALLEE_SAVE(native_irq_disable),
+	.irq_enable = __PV_IS_CALLEE_SAVE(native_irq_enable),
 	.safe_halt = native_safe_halt,
 	.halt = native_halt,
+#ifdef CONFIG_X86_64
+	.adjust_exception_frame = paravirt_nop,
+#endif
 };
 
 struct pv_cpu_ops pv_cpu_ops = {
@@ -369,11 +331,18 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.read_cr4 = native_read_cr4,
 	.read_cr4_safe = native_read_cr4_safe,
 	.write_cr4 = native_write_cr4,
+#ifdef CONFIG_X86_64
+	.read_cr8 = native_read_cr8,
+	.write_cr8 = native_write_cr8,
+#endif
 	.wbinvd = native_wbinvd,
 	.read_msr = native_read_msr_safe,
+	.rdmsr_regs = native_rdmsr_safe_regs,
 	.write_msr = native_write_msr_safe,
+	.wrmsr_regs = native_wrmsr_safe_regs,
 	.read_tsc = native_read_tsc,
 	.read_pmc = native_read_pmc,
+	.read_tscp = native_read_tscp,
 	.load_tr_desc = native_load_tr_desc,
 	.set_ldt = native_set_ldt,
 	.load_gdt = native_load_gdt,
@@ -382,37 +351,52 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.store_idt = native_store_idt,
 	.store_tr = native_store_tr,
 	.load_tls = native_load_tls,
+#ifdef CONFIG_X86_64
+	.load_gs_index = native_load_gs_index,
+#endif
 	.write_ldt_entry = native_write_ldt_entry,
 	.write_gdt_entry = native_write_gdt_entry,
 	.write_idt_entry = native_write_idt_entry,
+
+	.alloc_ldt = paravirt_nop,
+	.free_ldt = paravirt_nop,
+
 	.load_sp0 = native_load_sp0,
 
-	.irq_enable_syscall_ret = native_irq_enable_syscall_ret,
+#if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION)
+	.irq_enable_sysexit = native_irq_enable_sysexit,
+#endif
+#ifdef CONFIG_X86_64
+#ifdef CONFIG_IA32_EMULATION
+	.usergs_sysret32 = native_usergs_sysret32,
+#endif
+	.usergs_sysret64 = native_usergs_sysret64,
+#endif
 	.iret = native_iret,
+	.swapgs = native_swapgs,
 
 	.set_iopl_mask = native_set_iopl_mask,
 	.io_delay = native_io_delay,
 
-	.lazy_mode = {
-		.enter = paravirt_nop,
-		.leave = paravirt_nop,
-	},
+	.start_context_switch = paravirt_nop,
+	.end_context_switch = paravirt_nop,
 };
 
 struct pv_apic_ops pv_apic_ops = {
 #ifdef CONFIG_X86_LOCAL_APIC
-	.apic_write = native_apic_write,
-	.apic_write_atomic = native_apic_write_atomic,
-	.apic_read = native_apic_read,
-	.setup_boot_clock = setup_boot_APIC_clock,
-	.setup_secondary_clock = setup_secondary_APIC_clock,
 	.startup_ipi_hook = paravirt_nop,
 #endif
 };
 
+#if defined(CONFIG_X86_32) && !defined(CONFIG_X86_PAE)
+/* 32-bit pagetable entries */
+#define PTE_IDENT	__PV_IS_CALLEE_SAVE(_paravirt_ident_32)
+#else
+/* 64-bit pagetable entries */
+#define PTE_IDENT	__PV_IS_CALLEE_SAVE(_paravirt_ident_64)
+#endif
+
 struct pv_mmu_ops pv_mmu_ops = {
-	.pagetable_setup_start = native_pagetable_setup_start,
-	.pagetable_setup_done = native_pagetable_setup_done,
 
 	.read_cr2 = native_read_cr2,
 	.write_cr2 = native_write_cr2,
@@ -424,11 +408,16 @@ struct pv_mmu_ops pv_mmu_ops = {
 	.flush_tlb_single = native_flush_tlb_single,
 	.flush_tlb_others = native_flush_tlb_others,
 
-	.alloc_pt = paravirt_nop,
-	.alloc_pd = paravirt_nop,
-	.alloc_pd_clone = paravirt_nop,
-	.release_pt = paravirt_nop,
-	.release_pd = paravirt_nop,
+	.pgd_alloc = __paravirt_pgd_alloc,
+	.pgd_free = paravirt_nop,
+
+	.alloc_pte = paravirt_nop,
+	.alloc_pmd = paravirt_nop,
+	.alloc_pmd_clone = paravirt_nop,
+	.alloc_pud = paravirt_nop,
+	.release_pte = paravirt_nop,
+	.release_pmd = paravirt_nop,
+	.release_pud = paravirt_nop,
 
 	.set_pte = native_set_pte,
 	.set_pte_at = native_set_pte_at,
@@ -436,26 +425,37 @@ struct pv_mmu_ops pv_mmu_ops = {
 	.pte_update = paravirt_nop,
 	.pte_update_defer = paravirt_nop,
 
+	.ptep_modify_prot_start = __ptep_modify_prot_start,
+	.ptep_modify_prot_commit = __ptep_modify_prot_commit,
+
 #ifdef CONFIG_HIGHPTE
 	.kmap_atomic_pte = kmap_atomic,
 #endif
 
+#if PAGETABLE_LEVELS >= 3
 #ifdef CONFIG_X86_PAE
 	.set_pte_atomic = native_set_pte_atomic,
-	.set_pte_present = native_set_pte_present,
-	.set_pud = native_set_pud,
 	.pte_clear = native_pte_clear,
 	.pmd_clear = native_pmd_clear,
+#endif
+	.set_pud = native_set_pud,
+
+	.pmd_val = PTE_IDENT,
+	.make_pmd = PTE_IDENT,
 
-	.pmd_val = native_pmd_val,
-	.make_pmd = native_make_pmd,
+#if PAGETABLE_LEVELS == 4
+	.pud_val = PTE_IDENT,
+	.make_pud = PTE_IDENT,
+
+	.set_pgd = native_set_pgd,
 #endif
+#endif /* PAGETABLE_LEVELS >= 3 */
 
-	.pte_val = native_pte_val,
-	.pgd_val = native_pgd_val,
+	.pte_val = PTE_IDENT,
+	.pgd_val = PTE_IDENT,
 
-	.make_pte = native_make_pte,
-	.make_pgd = native_make_pgd,
+	.make_pte = PTE_IDENT,
+	.make_pgd = PTE_IDENT,
 
 	.dup_mmap = paravirt_nop,
 	.exit_mmap = paravirt_nop,
@@ -465,6 +465,8 @@ struct pv_mmu_ops pv_mmu_ops = {
 		.enter = paravirt_nop,
 		.leave = paravirt_nop,
 	},
+
+	.set_fixmap = native_set_fixmap,
 };
 
 EXPORT_SYMBOL_GPL(pv_time_ops);