1 files changed, 73 insertions, 86 deletions

diff --git a/arch/i386/kernel/vmi.c b/arch/i386/kernel/vmi.c
index fb07a1aad225..c8726c424b35 100644
--- a/arch/i386/kernel/vmi.c
+++ b/arch/i386/kernel/vmi.c
@@ -26,6 +26,7 @@
 #include <linux/cpu.h>
 #include <linux/bootmem.h>
 #include <linux/mm.h>
+#include <linux/highmem.h>
 #include <asm/vmi.h>
 #include <asm/io.h>
 #include <asm/fixmap.h>
@@ -56,7 +57,7 @@ static int disable_noidle;
 static int disable_vmi_timer;
 
 /* Cached VMI operations */
-struct {
+static struct {
 	void (*cpuid)(void /* non-c */);
 	void (*_set_ldt)(u32 selector);
 	void (*set_tr)(u32 selector);
@@ -65,15 +66,15 @@ struct {
 	void (*release_page)(u32, u32);
 	void (*set_pte)(pte_t, pte_t *, unsigned);
 	void (*update_pte)(pte_t *, unsigned);
-	void (*set_linear_mapping)(int, u32, u32, u32);
-	void (*flush_tlb)(int);
+	void (*set_linear_mapping)(int, void *, u32, u32);
+	void (*_flush_tlb)(int);
 	void (*set_initial_ap_state)(int, int);
 	void (*halt)(void);
+  	void (*set_lazy_mode)(int mode);
 } vmi_ops;
 
-/* XXX move this to alternative.h */
-extern struct paravirt_patch __start_parainstructions[],
-	__stop_parainstructions[];
+/* Cached VMI operations */
+struct vmi_timer_ops vmi_timer_ops;
 
 /*
  * VMI patching routines.
@@ -82,11 +83,6 @@ extern struct paravirt_patch __start_parainstructions[],
 #define MNEM_JMP  0xe9
 #define MNEM_RET  0xc3
 
-static char irq_save_disable_callout[] = {
-	MNEM_CALL, 0, 0, 0, 0,
-	MNEM_CALL, 0, 0, 0, 0,
-	MNEM_RET
-};
 #define IRQ_PATCH_INT_MASK 0
 #define IRQ_PATCH_DISABLE  5
 
@@ -134,33 +130,17 @@ static unsigned patch_internal(int call, unsigned len, void *insns)
 static unsigned vmi_patch(u8 type, u16 clobbers, void *insns, unsigned len)
 {
 	switch (type) {
-		case PARAVIRT_IRQ_DISABLE:
+		case PARAVIRT_PATCH(irq_disable):
 			return patch_internal(VMI_CALL_DisableInterrupts, len, insns);
-		case PARAVIRT_IRQ_ENABLE:
+		case PARAVIRT_PATCH(irq_enable):
 			return patch_internal(VMI_CALL_EnableInterrupts, len, insns);
-		case PARAVIRT_RESTORE_FLAGS:
+		case PARAVIRT_PATCH(restore_fl):
 			return patch_internal(VMI_CALL_SetInterruptMask, len, insns);
-		case PARAVIRT_SAVE_FLAGS:
+		case PARAVIRT_PATCH(save_fl):
 			return patch_internal(VMI_CALL_GetInterruptMask, len, insns);
-        	case PARAVIRT_SAVE_FLAGS_IRQ_DISABLE:
-			if (len >= 10) {
-				patch_internal(VMI_CALL_GetInterruptMask, len, insns);
-				patch_internal(VMI_CALL_DisableInterrupts, len-5, insns+5);
-				return 10;
-			} else {
-				/*
-				 * You bastards didn't leave enough room to
-				 * patch save_flags_irq_disable inline.  Patch
-				 * to a helper
-				 */
-				BUG_ON(len < 5);
-				*(char *)insns = MNEM_CALL;
-				patch_offset(insns, irq_save_disable_callout);
-				return 5;
-			}
-		case PARAVIRT_INTERRUPT_RETURN:
+		case PARAVIRT_PATCH(iret):
 			return patch_internal(VMI_CALL_IRET, len, insns);
-		case PARAVIRT_STI_SYSEXIT:
+		case PARAVIRT_PATCH(irq_enable_sysexit):
 			return patch_internal(VMI_CALL_SYSEXIT, len, insns);
 		default:
 			break;
@@ -229,24 +209,24 @@ static void vmi_set_tr(void)
 static void vmi_load_esp0(struct tss_struct *tss,
 				   struct thread_struct *thread)
 {
-	tss->esp0 = thread->esp0;
+	tss->x86_tss.esp0 = thread->esp0;
 
 	/* This can only happen when SEP is enabled, no need to test "SEP"arately */
-	if (unlikely(tss->ss1 != thread->sysenter_cs)) {
-		tss->ss1 = thread->sysenter_cs;
+	if (unlikely(tss->x86_tss.ss1 != thread->sysenter_cs)) {
+		tss->x86_tss.ss1 = thread->sysenter_cs;
 		wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
 	}
-	vmi_ops.set_kernel_stack(__KERNEL_DS, tss->esp0);
+	vmi_ops.set_kernel_stack(__KERNEL_DS, tss->x86_tss.esp0);
 }
 
 static void vmi_flush_tlb_user(void)
 {
-	vmi_ops.flush_tlb(VMI_FLUSH_TLB);
+	vmi_ops._flush_tlb(VMI_FLUSH_TLB);
 }
 
 static void vmi_flush_tlb_kernel(void)
 {
-	vmi_ops.flush_tlb(VMI_FLUSH_TLB | VMI_FLUSH_GLOBAL);
+	vmi_ops._flush_tlb(VMI_FLUSH_TLB | VMI_FLUSH_GLOBAL);
 }
 
 /* Stub to do nothing at all; used for delays and unimplemented calls */
@@ -254,18 +234,6 @@ static void vmi_nop(void)
 {
 }
 
-/* For NO_IDLE_HZ, we stop the clock when halting the kernel */
-static fastcall void vmi_safe_halt(void)
-{
-	int idle = vmi_stop_hz_timer();
-	vmi_ops.halt();
-	if (idle) {
-		local_irq_disable();
-		vmi_account_time_restart_hz_timer();
-		local_irq_enable();
-	}
-}
-
 #ifdef CONFIG_DEBUG_PAGE_TYPE
 
 #ifdef CONFIG_X86_PAE
@@ -369,8 +337,11 @@ static void vmi_check_page_type(u32 pfn, int type)
 #define vmi_check_page_type(p,t) do { } while (0)
 #endif
 
-static void vmi_map_pt_hook(int type, pte_t *va, u32 pfn)
+#ifdef CONFIG_HIGHPTE
+static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type)
 {
+	void *va = kmap_atomic(page, type);
+
 	/*
 	 * Internally, the VMI ROM must map virtual addresses to physical
 	 * addresses for processing MMU updates.  By the time MMU updates
@@ -384,8 +355,11 @@ static void vmi_map_pt_hook(int type, pte_t *va, u32 pfn)
 	 *  args:                 SLOT                 VA    COUNT PFN
 	 */
 	BUG_ON(type != KM_PTE0 && type != KM_PTE1);
-	vmi_ops.set_linear_mapping((type - KM_PTE0)+1, (u32)va, 1, pfn);
+	vmi_ops.set_linear_mapping((type - KM_PTE0)+1, va, 1, page_to_pfn(page));
+
+	return va;
 }
+#endif
 
 static void vmi_allocate_pt(u32 pfn)
 {
@@ -442,13 +416,13 @@ static void vmi_release_pd(u32 pfn)
         ((level) | (is_current_as(mm, user) ?                           \
                 (VMI_PAGE_DEFER | VMI_PAGE_CURRENT_AS | ((addr) & VMI_PAGE_VA_MASK)) : 0))
 
-static void vmi_update_pte(struct mm_struct *mm, u32 addr, pte_t *ptep)
+static void vmi_update_pte(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE);
 	vmi_ops.update_pte(ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
 }
 
-static void vmi_update_pte_defer(struct mm_struct *mm, u32 addr, pte_t *ptep)
+static void vmi_update_pte_defer(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE);
 	vmi_ops.update_pte(ptep, vmi_flags_addr_defer(mm, addr, VMI_PAGE_PT, 0));
@@ -461,7 +435,7 @@ static void vmi_set_pte(pte_t *ptep, pte_t pte)
 	vmi_ops.set_pte(pte, ptep, VMI_PAGE_PT);
 }
 
-static void vmi_set_pte_at(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pte)
+static void vmi_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte)
 {
 	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE);
 	vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
@@ -515,7 +489,7 @@ static void vmi_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 	vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
 }
 
-void vmi_pmd_clear(pmd_t *pmd)
+static void vmi_pmd_clear(pmd_t *pmd)
 {
 	const pte_t pte = { 0 };
 	vmi_check_page_type(__pa(pmd) >> PAGE_SHIFT, VMI_PAGE_PMD);
@@ -524,8 +498,6 @@ void vmi_pmd_clear(pmd_t *pmd)
 #endif
 
 #ifdef CONFIG_SMP
-extern void setup_pda(void);
-
 static void __devinit
 vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip,
 		     unsigned long start_esp)
@@ -550,13 +522,11 @@ vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip,
 
 	ap.ds = __USER_DS;
 	ap.es = __USER_DS;
-	ap.fs = __KERNEL_PDA;
+	ap.fs = __KERNEL_PERCPU;
 	ap.gs = 0;
 
 	ap.eflags = 0;
 
-	setup_pda();
-
 #ifdef CONFIG_X86_PAE
 	/* efer should match BSP efer. */
 	if (cpu_has_nx) {
@@ -574,6 +544,26 @@ vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip,
 }
 #endif
 
+static void vmi_set_lazy_mode(enum paravirt_lazy_mode mode)
+{
+	static DEFINE_PER_CPU(enum paravirt_lazy_mode, lazy_mode);
+
+	if (!vmi_ops.set_lazy_mode)
+		return;
+
+	/* Modes should never nest or overlap */
+	BUG_ON(__get_cpu_var(lazy_mode) && !(mode == PARAVIRT_LAZY_NONE ||
+					     mode == PARAVIRT_LAZY_FLUSH));
+
+	if (mode == PARAVIRT_LAZY_FLUSH) {
+		vmi_ops.set_lazy_mode(0);
+		vmi_ops.set_lazy_mode(__get_cpu_var(lazy_mode));
+	} else {
+		vmi_ops.set_lazy_mode(mode);
+		__get_cpu_var(lazy_mode) = mode;
+	}
+}
+
 static inline int __init check_vmi_rom(struct vrom_header *rom)
 {
 	struct pci_header *pci;
@@ -664,7 +654,7 @@ void vmi_bringup(void)
 {
  	/* We must establish the lowmem mapping for MMU ops to work */
 	if (vmi_ops.set_linear_mapping)
-		vmi_ops.set_linear_mapping(0, __PAGE_OFFSET, max_low_pfn, 0);
+		vmi_ops.set_linear_mapping(0, (void *)__PAGE_OFFSET, max_low_pfn, 0);
 }
 
 /*
@@ -691,11 +681,14 @@ static void *vmi_get_function(int vmicall)
 do {								\
 	reloc = call_vrom_long_func(vmi_rom, get_reloc,		\
 				    VMI_CALL_##vmicall);	\
-	if (rel->type != VMI_RELOCATION_NONE) {			\
-		BUG_ON(rel->type != VMI_RELOCATION_CALL_REL);	\
+	if (rel->type == VMI_RELOCATION_CALL_REL) 		\
 		paravirt_ops.opname = (void *)rel->eip;		\
-	} else if (rel->type == VMI_RELOCATION_NOP) 		\
+	else if (rel->type == VMI_RELOCATION_NOP) 		\
 		paravirt_ops.opname = (void *)vmi_nop;		\
+	else if (rel->type != VMI_RELOCATION_NONE)		\
+		printk(KERN_WARNING "VMI: Unknown relocation "	\
+				    "type %d for " #vmicall"\n",\
+					rel->type);		\
 } while (0)
 
 /*
@@ -716,7 +709,6 @@ do {								\
 	}							\
 } while (0)
 
-
 /*
  * Activate the VMI interface and switch into paravirtualized mode
  */
@@ -772,12 +764,6 @@ static inline int __init activate_vmi(void)
 	para_fill(irq_disable, DisableInterrupts);
 	para_fill(irq_enable, EnableInterrupts);
 
-	/* irq_save_disable !!! sheer pain */
-	patch_offset(&irq_save_disable_callout[IRQ_PATCH_INT_MASK],
-		     (char *)paravirt_ops.save_fl);
-	patch_offset(&irq_save_disable_callout[IRQ_PATCH_DISABLE],
-		     (char *)paravirt_ops.irq_disable);
-
 	para_fill(wbinvd, WBINVD);
 	para_fill(read_tsc, RDTSC);
 
@@ -804,11 +790,11 @@ static inline int __init activate_vmi(void)
 	para_wrap(load_esp0, vmi_load_esp0, set_kernel_stack, UpdateKernelStack);
 	para_fill(set_iopl_mask, SetIOPLMask);
 	para_fill(io_delay, IODelay);
-	para_fill(set_lazy_mode, SetLazyMode);
+	para_wrap(set_lazy_mode, vmi_set_lazy_mode, set_lazy_mode, SetLazyMode);
 
 	/* user and kernel flush are just handled with different flags to FlushTLB */
-	para_wrap(flush_tlb_user, vmi_flush_tlb_user, flush_tlb, FlushTLB);
-	para_wrap(flush_tlb_kernel, vmi_flush_tlb_kernel, flush_tlb, FlushTLB);
+	para_wrap(flush_tlb_user, vmi_flush_tlb_user, _flush_tlb, FlushTLB);
+	para_wrap(flush_tlb_kernel, vmi_flush_tlb_kernel, _flush_tlb, FlushTLB);
 	para_fill(flush_tlb_single, InvalPage);
 
 	/*
@@ -854,8 +840,13 @@ static inline int __init activate_vmi(void)
 		paravirt_ops.release_pt = vmi_release_pt;
 		paravirt_ops.release_pd = vmi_release_pd;
 	}
-	para_wrap(map_pt_hook, vmi_map_pt_hook, set_linear_mapping,
-		  SetLinearMapping);
+
+	/* Set linear is needed in all cases */
+	vmi_ops.set_linear_mapping = vmi_get_function(VMI_CALL_SetLinearMapping);
+#ifdef CONFIG_HIGHPTE
+	if (vmi_ops.set_linear_mapping)
+		paravirt_ops.kmap_atomic_pte = vmi_kmap_atomic_pte;
+#endif
 
 	/*
 	 * These MUST always be patched.  Don't support indirect jumps
@@ -896,8 +887,8 @@ static inline int __init activate_vmi(void)
 		paravirt_ops.get_wallclock = vmi_get_wallclock;
 		paravirt_ops.set_wallclock = vmi_set_wallclock;
 #ifdef CONFIG_X86_LOCAL_APIC
-		paravirt_ops.setup_boot_clock = vmi_timer_setup_boot_alarm;
-		paravirt_ops.setup_secondary_clock = vmi_timer_setup_secondary_alarm;
+		paravirt_ops.setup_boot_clock = vmi_time_bsp_init;
+		paravirt_ops.setup_secondary_clock = vmi_time_ap_init;
 #endif
 		paravirt_ops.get_scheduled_cycles = vmi_get_sched_cycles;
  		paravirt_ops.get_cpu_khz = vmi_cpu_khz;
@@ -909,11 +900,7 @@ static inline int __init activate_vmi(void)
 		disable_vmi_timer = 1;
 	}
 
-	/* No idle HZ mode only works if VMI timer and no idle is enabled */
-	if (disable_noidle || disable_vmi_timer)
-		para_fill(safe_halt, Halt);
-	else
-		para_wrap(safe_halt, vmi_safe_halt, halt, Halt);
+	para_fill(safe_halt, Halt);
 
 	/*
 	 * Alternative instruction rewriting doesn't happen soon enough
@@ -921,7 +908,7 @@ static inline int __init activate_vmi(void)
 	 * to do this before IRQs get reenabled.  Fortunately, it is
 	 * idempotent.
 	 */
-	apply_paravirt(__start_parainstructions, __stop_parainstructions);
+	apply_paravirt(__parainstructions, __parainstructions_end);
 
 	vmi_bringup();